LOS ANGELES — On Friday Dr. Deborah Birx, the U.S. coronavirus response coordinator, praised California and Washington state for their social distancing efforts, claiming that they may have already slowed the spread of the virus and should serve as models for the rest of the nation.
“We really do appreciate the work of the citizens of California and Washington state, because we do see that their curve is different,” Birx said at the daily White House briefing. “Their curve is different from New York, New Jersey and Connecticut — and we really believe that the work that every citizen is doing in those states is making a difference.”
Nationally, the numbers have been devastating. More than 273,000 confirmed COVID-19 cases across the United States, with a staggering 100,000 in New York alone. A death toll — roughly 7,000 at last count — that has already doubled 9/11’s. And grim new estimates from the Trump administration that predict the deadly pathogen could eventually kill between 100,000 and 240,000 Americans — assuming the U.S. does everything right.
Yet in recent days, a few bright spots have started to flicker amid the statistical gloom. They come from various sources: internet-connected thermometers; smartphone GPS data; a study by private researchers in Bellevue, Wash.; state-by-state projections from the University of Washington; reports from hospitals in the San Francisco Bay Area.
All of this new information points to the same hopeful, if tentative, conclusion. In areas of America that caught the coronavirus early in its trajectory — and quickly implemented the sort of strict, sweeping social distancing measures recommended by public health officials — the curve may be starting to bend. The virus may be starting to spread more slowly. Staying home may be starting to work.
And as Birx has been telling the public all week, the two states to watch are Washington and California. For now, only 8 percent of COVID-19 tests there are coming back positive — compared to 35 percent in New York and New Jersey, 26 percent in Louisiana and 15 percent in Michigan, Connecticut, Indiana, Georgia and Illinois.
“California and Washington reacted very early to this,” Birx told reporters Tuesday. “Washington state, early, about two weeks before New York or New Jersey ... California, a week before New York or New Jersey, really talked to their communities and decided to mitigate before they started seeing this number of cases. And now we know that makes a big difference.”
The first known COVID-19 case in the U.S. was identified near Seattle, and it was there that the virus claimed 37 of its first 50 U.S. victims. After discovering in late February that residents with no known exposure or history of foreign travel were becoming infected, officials began to urge social distancing. Within a week, people were starting to work from home and seniors were being encouraged to stay indoors. On March 11, Gov. Jay Inslee banned large events in three Seattle-area counties and the city’s public schools announced they would close.
Five days later, the Bay Area stunned the nation when health officers in six counties jointly ordered residents to shelter in place. Eleven other California counties soon joined the order, and on March 19 it was expanded statewide by Gov. Gavin Newsom.
In contrast, it took New York state until March 22 to implement a similar order. Playgrounds in New York City didn’t close until Wednesday.
“People should go out and continue to live life, should go out to restaurants,” New York Mayor Bill de Blasio said on March 11.
A new “social distancing scoreboard” by New York technology company Unacast suggests that people are now following these shelter-in-place orders. Using smartphone GPS data to compare how far residents of every U.S. county travel each day, the scorecard awards New York, California and Washington high grades for cutting average mobility by more than 40 percent since the start of the outbreak. In fact, mobility has declined by more in Manhattan (72 percent) than in San Francisco (65 percent), Seattle (56 percent) or Los Angeles (46 percent).
But timing is everything during a pandemic, and it appears the virus had already spread widely in the nation’s largest, most densely populated city before the lockdown. Kinsa Health, which produces internet-connected thermometers, has been compiling and analyzing up to 162,000 daily temperature readings since COVID-19 struck the U.S., as the New York Times recently reported, adapting its software to detect spikes of “atypical fever” that do not correlate with historical flu patterns and are likely attributable to the coronavirus.
According to Kinsa’s national map, overall fevers have fallen or held steady in every county in the country over the last week — a sign that stay-at-home measures, which now cover more than 250 million Americans, are having an effect. Yet in both Manhattan and Brooklyn, the level of atypical illness over the last month registered as “high,” likely because so many people were already infected (and infectious) prior to social distancing. Meanwhile, the level of atypical illness was either low or mild in the Bay Area. Around Seattle, the level was either mild or moderate. And in Los Angeles, Kinsa lists the cumulative level of atypical illness as “none.”
Since 2018, Kinsa’s predictions have routinely been two to three weeks ahead of those of the Centers for Disease Control and Prevention, which relies on data from doctor office visits instead. Kinsa’s listing doesn’t mean no one in L.A. has COVID-19; according to the latest numbers from the Los Angeles Times, L.A. County has logged more than 3,000 cases and 50 deaths so far, and both of those numbers will continue to grow. But because there’s a lag between the time someone catches the virus and starts showing up in the data — it can take up to two weeks to experience symptoms, up to 10 more days before hospitalization is required and another two or three weeks before the case resolves, with either recovery or death — a lower level of atypical illness in a region that started sheltering in place several weeks ago suggests there may be fewer pre-social-distancing cases lurking in the pipeline. It also suggests that when the peak does arrive, it will likely be lower than New York’s.
This hypothesis may be starting to bear out. Preliminary statistical models provided to public officials in Washington state by the Institute for Disease Modeling, a private research group in Bellevue, suggest that the spread of the virus has slowed in the Seattle area in recent days. One month ago, each infected person was spreading the virus to an average of 2.7 other people; now that number appears to have been cut roughly in half, to 1.4.
Meanwhile, “the number of new cases diagnosed today in San Francisco is about two-thirds of what it was four or five days ago, which means the slope of the increase is lower than it was, which, to me, is reasonably good evidence of some flattening of the curve,” Dr. Bob Wachter, a professor and chair of the Department of Medicine at the University of California, San Francisco, told the New York Times on Tuesday. “If this thing was expanding in the way epidemics do when they’re not being mitigated or addressed, the growth [would be] exponential.”
As a result, the death toll has been doubling every seven or eight days in Washington and every four or five days in California — far slower than in New York, where it’s been doubling every two or three days. (Deaths are not a perfect measure of the virus’s spread. But they’re a far better way to compare places than confirmed cases, which reflect different levels of testing throughout the country.)
Cases will continue to rise in California and Washington, and the death toll will continue to climb. Things will get worse before they get better. But the situation on the West Coast will likely not be as bad as it would have otherwise been — or as bad as it still could be elsewhere. According to new projections, the Institute for Health Metrics and Evaluation at the University of Washington estimates that as many as 177,000 Americans may die from COVID-19 by August (again, if the country does everything right). Yet the same model says that only 11,000 of those deaths would occur in California, America’s most populous state, and that only 2,500 of them would occur in Washington (the site of America’s first outbreak). New York, by comparison, is on track to suffer 22,000 coronavirus deaths — twice as many as California and nearly 10 times as many as Washington.
Assuming its model is correct, the IHME also projects that neither California nor Washington is likely to exceed its hospital resource capacity. New York, by contrast, has already exceeded it, and is likely to continue to need many more hospital beds, intensive care beds and ventilators than it had before the pandemic.
“There’s an indication that maybe all the sheltering in place and doing that early helped bend the curve here, and it might not be as bad,” Eric Holdeman, the former director of King County (Seattle) disaster management, recently explained. “And maybe we can avoid what New York is going to experience in the next 10 days to two weeks, which is going to look like Italy and look like Spain.”
The takeaway for the rest of the country is clear: Move fast, act decisively and don’t let up. (Government models released Wednesday showed that as many as 5,000 Californians could die per week if restrictions are relaxed too soon.) Several large states, including Texas, Georgia, Florida and Pennsylvania, had yet to issue statewide shelter-in-place orders as recently as Tuesday — even though deaths are doubling faster there than in California. The death tolls in Ohio, Illinois, Michigan, New Jersey and Louisiana are rising at similarly rapid rates.
In some places — such as the emerging hot spots of Detroit, Chicago, Miami, Atlanta and New Orleans — it may be too late to prevent hospitals from being overwhelmed. Everywhere else should heed the lesson of California and Washington.
“If you wait till you see it,” Birx said Tuesday, “it’s too late.”
Click here for the latest coronavirus news and updates. According to experts, people over 60 and those who are immunocompromised continue to be the most at risk. If you have questions, please reference the CDC and WHO’s resource guides.