With exponential growth, the starting point matters, too. Some states had never fully locked down. And in two-thirds of states, cases were rising when governors began to ease their lockdowns. Georgia, for instance, reopened its economy—including businesses such as nail salons, gyms, and tattoo parlors—when daily new cases were averaging around seventy-five per million. (As The Economist pointed out, “By contrast, France, Spain and Italy had 13-17 new cases per million when they began to reopen their economies.”) In Texas, Governor Greg Abbott allowed bars to open on May 18th, a day in which there was an increase of nearly a thousand new cases, following weeks of a steady rise. In Florida, Disney World opened on July 11th, the day after the state recorded more than ten thousand new cases—and a single-day record of four hundred and thirty-five new hospitalizations.
Some specific factors, beyond the virus’s exponential growth, contributed to the summer’s surge in cases. In general, the early shortages of masks—and the confusion and conflicting political messages swirling around their use—delayed their widespread adoption. Last month, Georgia’s governor, for instance, sued the mayor of Atlanta when she ordered people to wear masks. President Trump wore a mask publicly for the first time in late July. But evidence now shows that masks not only help prevent asymptomatic infected individuals from spreading the virus but also might protect wearers themselves against severe or even moderate illness, by reducing the viral dose—the amount of virus that a person initially inhales. (The greater the dose, the harder it becomes for the immune system to fight off infection.) According to a recent paper, eighty-one per cent of the infected passengers and staff on a cruise ship—all of whom had been issued masks—remained asymptomatic, and, in a seafood factory in Oregon, where workers received masks each day, the rate of asymptomatic infection among the hundred and twenty-four infected was ninety-five per cent. Monica Gandhi, one of the paper’s co-authors and an infectious-disease specialist at the University of California, San Francisco, told NPR, “It really is that the less virus that you get in, the less sick you’re likely to be.”
Bars, house parties, and other gatherings in May and June also likely helped fuel the spread. “When we talk about the reopening, it’s important to remember why we closed,” Emily Gurley, the epidemiologist, said. The first, explicit reason was, obviously, to bring transmission down—or, in other words, to flatten the curve. “Why does closing things down work? Because no one has contacts anymore,” Gurley said. A second, implicit reason was to buy time to establish the necessary public-health infrastructure. “The virus didn’t go away. You’re still on an incline,” Gurley said. “We took our foot off the brakes before we had anything else to catch us.”
Since early summer, ten states and Puerto Rico have reimposed some of their earlier restrictions, and thirteen have paused their reopenings. Since May, several states have made an effort to ramp up their contact-tracing efforts, including California, which has hired and trained more than seven thousand disease detectives over the past two months, and is attempting to get up to twenty thousand. On June 1st, New York City launched its Test & Trace Corps, hiring and training three thousand disease detectives—up from about a hundred in April. But over all, the White House still has not made it a national priority. “What we need,” Crystal Watson, the contact-tracing expert, told me, “is a call to action at a national level and much more federal funding.”
Even though testing has been made a priority in the national response, in many regions, especially those with the largest case surges, there are still long delays in turning around results, as much as a week to ten days. “I can’t tell you how frustrating it is to be back in the same place,” Gurley said. “We’ve got to be screaming about this again.” With test-result delays, contact tracing becomes somewhat futile, since the point is to reach an infected person’s possible contacts and ask them to quarantine before they start shedding virus. People, on average, pose an infectious risk roughly two days before symptoms start. “Contact tracing is one of the best tools we have for being able to identify as many people as possible who have COVID, even in the absence of symptoms,” Mike Reid, the chief medical officer for California’s contact-tracer training effort, said. Watson expressed concern that because some states will now not be able to control surging outbreaks with contact tracing alone, “governors will give up on developing this capacity,” even though it will be essential to prevent future waves of infection.
Gurley pointed out that what we don’t know about transmission, at this point, does not matter much in terms of policy. “The U.S. has been focussed on a few trees we were unsure about and lost view of the forest of everything we do know,” she said. “Most people are infected by someone they live with, someone they had very close contact with, and most transmission happens from people who are sick or going to be sick.” Lessler told me that people are, in fact, five to ten times more likely to be infected by a close household contact than any other kind of contact. One new study, based on data from South Korea, where officials tracked down nearly every contact of anyone who was sick, found that the majority of contacts—fifty-seven per cent—who tested positive for COVID-19 were household contacts, versus contacts from a restaurant or store. “Even if we just focussed on people that are easy to find,” Gurley said, “we could have a huge impact.”
If, contrary to the prevailing data, I had not picked up the virus before I left my temporary home in Brooklyn, there are only four other possible places where I might have been exposed. Right before I arrived at the Long Island house, on April 1st, I stopped at a roadside restaurant for takeout. There was no one else in the parking lot, and two employees were inside. One of them, a tall, burly, bearded man, handed off my food in a bag. I have since called the restaurant; the only employee who got sick was quarantined before my visit. (On a list of thirty-seven activities labelled low-risk to high-risk, which was released by the Texas Medical Association, getting takeout is firmly low-risk, labelled a two, on a scale from one to ten.)
Once I got to the house, I did not leave again (except to go for a walk or run along an empty path) until April 10th, when I took the trash to the town dump. The sorting center has a roof, but otherwise it’s open-air. Everyone I passed was wearing a mask. The next day, I took a box to the local U.P.S. store. I used a sleeve to open the door. There were two women working behind the counter, wearing masks. I wore a kerchief across my face and did not touch anything. The clerk even asked me to read her my credit-card information instead of passing her the card. Of the three other people who were in the store while I was there, one young woman was properly masked, one elderly man wore a mask but left his nose exposed, and one middle-aged woman, who was there only for a moment, wore her mask like a beard. Maybe I walked into a virus-contaminated cloud while inside, and my kerchief was insufficient protection. On April 12th, Easter Sunday, I went for a run on the beach. A big dog came running up to greet me. Instinctively, I scratched its head. I did not pay attention to who the owners were and, in any case, I never got near them. That’s it. No other possibilities. By the 14th, I was sick.