The country is reopening. What does the future hold?

The story of the American pandemic has unfolded in three chapters. The first began last January, when the coronavirus emerged and the world was plunged into uncertainty about how covid-19 could be treated, how the virus spread, and when it might be defeated. The second started on the morning of November 9, 2020, when Pfizer-BioNTech announced the extraordinary efficacy of its vaccine. Those results made clear that this pandemic would end not through infection but vaccination. Our goals shifted from merely slowing the spread to beginning immunization as quickly as possible. In America, much of the past half year has been devoted to administering vaccines and gathering evidence on how well they work in the real world.

Earlier this month, the Centers for Disease Control and Prevention ushered in the American pandemic’s third chapter. The agency announced that vaccinated people could go without masks or social distancing indoors and outside, in crowds large and small. It carved out a few exceptions—for hospitals, public transportation, and the like—and noted that people still needed to obey federal and local laws. But the broad message was that vaccinated Americans could resume their pre-pandemic lives. The C.D.C. is an agency known for caution, and its new guidance shocked many public-health experts; just two weeks earlier, it had issued far more restrictive recommendations. During the same period, a survey of nearly six hundred epidemiologists found that more than three-quarters of them believed that indoor mask-wearing might remain necessary for another year or more. Still, immediately after the announcement, a number of states lifted their mask mandates. Others will surely follow, as the pressure to return to normal grows. America is now moving swiftly toward reopening.

Despite the C.D.C.’s early stumbles on communication, masks, and tests, it remains perhaps the world’s preëminent public-health agency. Its recommendations carry unparalleled scientific force in the U.S. and beyond. Ultimately, the C.D.C.’s decision reflects real shifts in the weight of the evidence on several fundamental epidemiological questions: Are the vaccines as effective as they were in the trials? Can they protect us against the coronavirus variants? And do they prevent not just illness but transmission? The answers to these questions give us good reason to think that the pandemic’s newest chapter will be its last. Read The New Yorker’s complete news coverage and analysis of the coronavirus pandemic.

On the first question, the nationwide rollout of covid-19 vaccines has proved, beyond any doubt, that they are astonishingly effective at preventing serious illness, even for the most vulnerable people. So-called breakthrough infections, in which the virus weaves its way around some of an individual’s immune system, do occur. But such infections are extremely rare, and—because a person almost always has some effective antibodies and other immune-system defenses—they usually cause mild or no symptoms. In one study, the C.D.C. examined post-vaccination infections among nearly fifteen thousand nursing-home residents and staff members, and discovered only two covid-19 hospitalizations and one death. Another study, involving half a million health-care workers from around the country, found that getting two shots reduced the risk of a symptomatic infection by ninety-four per cent. Moving forward, we should expect to continue seeing breakthrough infections from time to time—but, for the most part, we shouldn’t worry about them. (At the same time, the covid vaccines have proved exceptionally safe. Few dangerous side effects have been linked to the vaccines from Pfizer-BioNTech or Moderna, and the over-all risk of concerning blood clots after receiving Johnson & Johnson’s vaccine is rare—as of last week, when more than nine million doses had been administered, there were thirty confirmed cases.)

The most striking vaccine-efficacy statistic draws on data shared by state governments. Around a hundred and thirty million Americans are fully vaccinated, and the C.D.C. has said that it has received reports of fewer than fourteen hundred covid-19 hospitalizations and three hundred deaths among them. This means that, after vaccination, one’s chances of dying of covid-19 are currently about two in a million, with the likelihood of being hospitalized only slightly higher. Statistics reported by hospitals tend to be accurate; still, even if state governments have missed a few cases here and there, the results are staggeringly good. “The evidence on vaccines just keeps getting better and better,” Robert Wachter, a physician and the chair of medicine at the University of California, San Francisco, told me. “When the trial results first came out, I thought, They can’t actually be this good. The real world is always messier than the trials. What we’ve learned since then is that the vaccines are probably even more spectacular than we initially believed.”

The answer to the second question—whether the vaccines work against the major coronavirus variants—is also now clear. Earlier this month, a study conducted in Qatar, where the B.1.1.7 and B.1.351 variants predominate, found that the Pfizer-BioNTech vaccine was ninety-seven per cent effective at preventing severe disease. Vaccines from Moderna and Johnson & Johnson also appear to be highly effective against the variants; in fact, these vaccines are already successfully fighting them here in the United States. The B.1.1.7 variant, which is vastly more contagious than the original virus and caused a devastating surge in the U.K. this past winter, now accounts for three-quarters of new U.S. cases—and yet, largely thanks to vaccination, daily infections in this country have fallen by nearly ninety per cent since their peak in January, and are now lower than at any point in the past eight months. The existence of more contagious variants isn’t a reason to doubt the vaccines but to vaccinate people as quickly as possible.

As for the final question—whether vaccinated people can spread the virus to others, especially unvaccinated people, including children—the evidence is similarly encouraging. Because vaccinated people are unlikely to contract the virus, the vast majority won’t be passing it on. And even the small number of vaccinated people who experience breakthrough infections have much less of the virus circulating in their bodies, and may be less infectious. Real-world data from Israel, which has mounted one of the world’s fastest and most effective vaccination campaigns, is instructive. The country’s progress in immunizing its adults has been linked to significant declines in infections among unvaccinated people; according to one preliminary estimate, each twenty-percentage-point increase in adult vaccination rates reduces infections for unvaccinated children by half. When vaccinated people remove their masks, they pose little threat to others, and they face little peril themselves.

The shift toward reopening is not without risk. The first issue is timing. Less than half of Americans have received even one shot of a covid-19 vaccine, and only around four in ten have been fully vaccinated. This means that the majority of the country remains susceptible to infection and disease. Meanwhile, the pace of vaccinations has slowed: in April, the U.S. was routinely vaccinating about three million people per day, but the daily average is now nearly two million. It’s unclear whether the new guidance will encourage or deter unvaccinated Americans from getting immunized. In a recent survey, unvaccinated Republicans said that they would be nearly twenty per cent more likely to get the shots if it meant that they wouldn’t have to wear a mask anymore. We’ll now find out how they really feel.

Vaccine hesitancy is only part of the picture. Some thirty million Americans—a group larger than anti-vaxxers or the vaccine-hesitant—say that they want to get immunized but haven’t yet done so. Some face language barriers, or fear immigration problems; others have difficulty navigating the health system, or can’t take time off from work. Many of the willing-but-unvaccinated are working-class Americans; four in five don’t have a college degree. The Biden Administration has sent billions of dollars to health centers serving low-income populations, offered tax credits to businesses that provide paid time off for employees to get immunized, and helped assemble thousands of volunteers—known as the covid-19 Community Corps—to assist with vaccine outreach to underserved populations. States, too, are trying to reduce barriers to vaccination, and offering incentives—including payments in Maryland, a lottery in Ohio, and a “Shot and a Beer” program in New Jersey—for residents who remain on the fence. There are, in short, real efforts under way to sway the vaccine-hesitant and make vaccines more accessible.

Still, the new C.D.C. guidance makes these efforts even more urgent. Until now, unvaccinated people have been shielded from high levels of viral exposure by government mandates and social norms that have kept their friends, neighbors, and colleagues masked and distanced, to varying degrees. But, in the coming weeks, those protections will likely erode. For unvaccinated Americans, this could be the most dangerous moment in the pandemic. In most contexts, there is no reliable mechanism for verifying who has and hasn’t been vaccinated. Inevitably, against the C.D.C.’s advice, many unvaccinated people will resume normal life, too, threatening their own health and that of others. When asked how businesses are to know which customers can enter unmasked, Anthony Fauci, the nation’s top infectious-disease expert, told CNN, “They will not be able to know. You’re going to be depending on people being honest enough to say whether they were vaccinated or not.”

“Unvaccinated people are now going to have much higher levels of exposure,” Wachter told me. “That’s especially true in places with lots of community spread and in places where more contagious variants are circulating.” Wachter suggested that the C.D.C. could be making an epidemiological bet. The move “will cause some additional covid cases that otherwise would not have occurred,” he said—but, “if it leads to even a small uptick in vaccination, it will save lives in aggregate.”

Since the start of the pandemic’s second chapter, public-health officials have been working to prevent a catastrophic collision between the ship of reopening and the iceberg of the unvaccinated. By slowing the speed of the ship or shrinking the size of the iceberg, we have sought to reduce the force of the collision. But barring a hundred-per-cent vaccination rate, or something close to it—an outcome that the U.S. was never likely to achieve—a crash of some sort has been inevitable. India’s collision has been titanic—it reopened with a population of more than a billion, even though hardly anyone was vaccinated. In the U.S., the situation is different. Our iceberg has been melting, and we’ve been approaching it slowly. Now we’re taking off the brakes.

The C.D.C. issues guidance, not laws; there are several quantitative measures that states, counties, cities, companies, and individuals can consult in pacing their reopening and squaring the agency’s broad recommendations with local realities. A community’s immunization rate is perhaps the most obvious statistic to track. Experts have argued for meeting a seventy-per-cent immunity threshold before relaxing masking and distancing requirements. No states have got there yet, although some, such as Vermont and Maine, are well on their way. The Biden Administration has said that it hopes to hit the seventy-per-cent target for first shots by the Fourth of July.

Because the vaccines prevent almost all cases of severe covid-19, the number of covid-19 hospitalizations is another good metric to watch. “With vaccines, cases become uncoupled from severe disease,” Monica Gandhi, an infectious-disease doctor at the University of California, San Francisco, who has studied asymptomatic coronavirus transmission, told me. Gandhi was among the first researchers to show that masks protect not just others but wearers, too; when we spoke, before the C.D.C.’s announcement, she said that, in her view, most precautions could end when half of Americans had received their first shot and covid-19 hospitalizations had fallen below sixteen thousand nationally, or about five per hundred thousand people. (At the peak of most flu seasons, the U.S. records five to ten influenza hospitalizations per hundred thousand.) Hospitalizations appear to be falling, unevenly, across the country. However, there are currently thirty thousand Americans hospitalized with covid-19—roughly a quarter of the January peak, but still about twice Gandhi’s threshold.

Herd immunity offers a third benchmark for reopening. The idea is that, once about eighty per cent of the population has been vaccinated or infected, the virus will struggle to spread. Recently, some experts have argued that we might never get to herd immunity because of variants, vaccine hesitancy, and the fact that children under twelve, who make up some fifteen per cent of the U.S. population, are unlikely to be immunized for some time. But the C.D.C.’s recommendation could change the equation. As states lift restrictions and unvaccinated people face higher levels of exposure, more of them are likely to get infected, pushing us closer to the herd immunity threshold. In all likelihood, the U.S. will be able to reach sixty-per-cent vaccination in the coming weeks; meanwhile, perhaps a third of Americans have already been infected. Even assuming significant overlap between the two groups, the combination of vaccination and infection is likely to make it harder for the virus to find new hosts. Marc Lipsitch, the director of Harvard’s Center for Communicable Disease Dynamics, emphasized that, because some parts of the country may reach herd immunity, or something close to it, before others—Connecticut’s current covid-19 immunization rate, for instance, is nearly twice Mississippi’s—unvaccinated adults will face different levels of risk depending on where they live. “There won’t be one national end,” Lipsitch told me. “We’re going to see a fundamental change in terms of what it means to live in this country, but there’s also going to be a lot of local variation.”

Covid-19 deaths give us another way of tracking the pandemic. Experts have argued that the U.S., with a population of three hundred and thirty-two million, should aim for fewer than a hundred coronavirus deaths daily—roughly the toll of a typical flu season. Right now, America is seeing about six hundred covid-19 deaths each day; according to the Institute for Health Metrics and Evaluation, which generates one of the country’s most widely cited pandemic models, that number will likely fall to about a hundred in August. “Things will look very good this summer,” Christopher Murray, the director of the I.H.M.E., told me. “A lot of people will think that we’re done, that it’s all over. But what happens in the fall is the tricky part.” Murray believes that a confluence of factors—the spread of variants, in-person schooling, meaningful numbers of still-unvaccinated people, and the seasonality of the virus—will produce a small winter spike, concentrated in communities with low vaccination rates. It won’t be the apocalyptic surge of New York City in the spring of 2020—or, more recently, those of India or Brazil—but, each week, several thousand unvaccinated Americans could die.

It’s possible, given all this, to imagine a plausible scenario for the conclusion of the American pandemic. The coronavirus disease toll continues to fall throughout the summer. States do away with mask mandates and capacity restrictions; people increasingly return to bars, spin classes, and airports, then to stadiums, movie theatres, and concerts. By midsummer, in communities with high vaccination rates, covid-19 starts to fade from view. In those places, even people who remain unvaccinated are protected, because so little of the virus circulates. But, in other parts of the country, low immunization rates combined with reopening allow the disease to register again. Hospitals aren’t overwhelmed—there’s no need to build new I.C.U.s or call in extra staff—but the collision between ship and iceberg is forceful, and each week thousands of people fall ill and hundreds die. Some victims are vaccine-hesitant; others were unable, for whatever reason, to get vaccinated. Still, perhaps unfairly, these outbreaks come with an aura of culpability: to people in safe parts of the country, the ill seem like smokers who get lung cancer.

In the fall, many unvaccinated children return to school. Scattered infections among them capture headlines, but serious illnesses are exceedingly rare; the overwhelming majority of children remain safe, and, with time, they, too, are immunized. The U.S. approaches something like herd immunity. Some people may still fall ill and die of covid-19—perhaps they are immunocompromised, elderly, or just unlucky—but, by and large, America has gained the upper hand. Meanwhile, in poor nations with few vaccines, the pandemic continues. As crisis wanes in one country, catastrophe ignites in another. Every so often, we learn of a new variant that’s thought to be more contagious, lethal, or vaccine-resistant than the rest; we rush to institute travel bans, only to learn that the variant, or a close cousin, is already circulating in the U.S. and has been largely subdued by the vaccines, as all previous variants have been. In the fall, Americans line up for covid booster shots alongside flu vaccines. The pandemic’s final chapter comes to a close not through official decree but with the gradual realization that covid-19 no longer dominates our lives.

Reopening a country after a pandemic isn’t like flipping a giant switch. It’s more like lighting a series of candles, illuminating one part, then another, until the whole place shines. Many states, counties, cities, and businesses will further loosen their restrictions; others will wait. Communities and individuals will approach the end of the crisis differently, as they’ve approached the rest of it. Some unvaccinated people have already been forgoing precautions; on the other hand, I’ve been vaccinated for months and, since the C.D.C. announcement, have yet to leave my mask behind—whether because of a lingering, irrational fear or simply to avoid dirty looks, I can’t say. Social norms take time to change, even when one of the world’s most respected public-health agencies is telling you to change them.

The pandemic has created not just chaos and suffering but uncertainty. It’s easy, therefore, to be doubtful about the fortunate position in which we seem to find ourselves now. As a physician, I spent the early months of the pandemic caring for covid-19 patients in New York City; they streamed into the hospital day after day, deathly ill. We raced to build covid wards, I.C.U.s, and hospice units. At the time, we had little to offer. There were no proven therapies, and certainly no vaccines. There were weeks when thousands of New Yorkers died, many of them alone in their final moments, while more people were dying across the world. I felt fear, anxiety, and sometimes despair. The scale of the damage—the lives lost, businesses shuttered, dreams shattered, children orphaned, seniors isolated—was crushing, and the path forward was both frightening and unknown.

As good news began to arrive, I greeted it with a blend of guarded skepticism and cautious optimism. First came evidence that outdoor transmission was unlikely. Then we learned that contaminated surfaces rarely spread disease; that some patients can breathe better simply by lying on their bellies; that P.P.E. works; that dexamethasone saves lives. We discovered that immunity lasts many months, perhaps years; that repeat infections are unlikely; and that variants present a surmountable challenge.

Now, study after study, in country after country, has shown that the vaccines are capable of transforming a lethal pathogen into a manageable threat. Examining and reëxamining the vaccine results, I’ve gone through stages, too—caution, hope, and, finally, clarity. We really are that close. The beginning of the end is here.