It began as a mysterious disease with frightening potential. Now, just two months after America's first confirmed case, the country is grappling with a lethal reality: The coronavirus has killed more than 1,000 people in the United States, a toll that is increasing at an alarming rate.
As the highly contagious virus has created clusters of illness, from Seattle to New York City, death has followed in turn. On Wednesday night, the country's largest city reported 88 new deaths from covid-19. As of Thursday afternoon, Americans had died in 42 states and territories and the District of Columbia, with punishing increases in Louisiana and Michigan. Experts fear the worst is still to come, pointing to a rapid acceleration of cases in communities across the country.
The Washington Post is tracking every known U.S. death, analyzing data from health agencies and gathering details from family and friends of the victims. In the first 1,000 fatalities, some patterns have begun to emerge in the outbreak's epidemiology and its painful human impact. About 65 percent of the dead whose ages are known were older than 70 and nearly 40 percent were over 80, demonstrating that risk rises along with age. About 5 percent whose ages are known were in their 40s or younger, but many more in that age group have been sick enough to be hospitalized. Of those victims whose gender is known, nearly 60 percent were men.
What remains murky is exactly who is dying in America during the pandemic, even as scientists and public health experts race to uncover information that can help save lives.
Overwhelmed state and local authorities have been issuing widely varying reports on those who died, citing medical privacy laws to shield even basic details about age, gender and underlying conditions, the three signal categories that epidemiologists say are key indicators of risk.
The Centers for Disease Control and Prevention, which offers a well-regarded and oft-cited public weekly tracker for the annual influenza season, offers no similar real-time surveillance for the novel coronavirus. The analysis the agency does provide relies on spotty reporting by the states, struggling to serve a surge of sick people.
There are some among the 1,000 deaths who publicly have a name, an age, a place of death and a life story: the playwright Terrence McNally, the rabbi Romi Cohn, the principal Dezann Romain. They appear in local media accounts: Sundee Rutter, a mother of six and breast cancer survivor in Washington state; Alvin Simmons, a father of two and hospital worker in New York; and Elizabeth Eugenia Wells, a grandmother who sang in her church choir in Georgia.
But many others surface only elliptically, in tweets and Facebook posts.
"Today, one of my friends died, presumably of complications related to coronavirus," a woman in Ohio posted in a tearful Facebook video message Monday. "She was my age. She had a husband, a daughter who is like 3. She was immunocompromised. She had some long-standing health problems. She wasn't, like, infirm. She was young and happy and vibrant. And now she's dead."
And: "My uncle died of Covid-19 today in California. He was 78 years old & had Parkinson's Disease. Went on some cruise prior to all the warnings. Went home, got ill, went to ER. Admitted. Lungs failed. ICU. Kidneys failed. Opted to DNR per MD 100% mortality rate opinion when pressed."
What becomes public varies widely by locality. In King County, Wash., an early epicenter of the U.S. outbreak, the health department posts daily updates to its website that include a victim's gender, age range, date of death, any presence of underlying health conditions and the hospital where they were treated. That report may include a note that the death is part of a cluster, such as at Life Care Center nursing home in Kirkland.
"Of the 87 deaths reported, 37 are confirmed to be associated with Life Care," King County reported on its site on Monday.
Some states, such as Florida and Colorado, offer detailed statewide data, while others, such as California, release almost no demographic details and defer to counties.
The inconsistency in reporting is particularly stark in New York. State health officials there have been taciturn about death statistics, usually leaving it to Gov. Andrew Cuomo, a Democrat, to announce the running toll at his daily briefings. In New York City, the health department has started releasing reports every day summarizing deaths by age group, gender, borough and preexisting medical problem.
Within them are numbers that raise red flags: data that shows at least 96 percent of those who died as of Wednesday had underlying conditions, and that 72 New Yorkers who died were under 64.
Still more deaths are not being counted at all, such as those who were misdiagnosed with the flu or another illness and those who died but were never tested, highlighting another key gap in mortality information.
One epidemiologist who left New York state's health department late last year said former colleagues have reached out to him recently, desperate to develop tools that would make it easier to track the disease as it spreads and kills.
"They not only lacked the ability to perform any type of modeling, forecasting or time series but could not even provide or perform basic epidemiologic analysis due to lack of access to data," the epidemiologist said, speaking on the condition of anonymity to candidly discuss operations within his field.
Ordinarily, experts say, the public could rest assured that the CDC is at least compiling detailed, nationwide data on the deaths and cases to analyze internally.
When a disease is categorized as "immediately notifiable, extremely urgent," as covid-19 is, officials are required to call and notify the CDC within hours of identifying a case.
"That means probable, suspected and then confirmed" cases of covid-19 - as well as deaths, said Charles Branas, the chair of the epidemiology department at Columbia University's Mailman School of Public Health. That call is then to be followed, within a day, with the submission of an electronic form, Branas said.
But state health departments - including, critically, New York's - are short-staffed and so deluged by the pandemic that they have not been filling out the forms with the basic information the CDC requires to perform an analysis, CDC officials say.
The New York State Department of Health even recently solicited volunteer help from local public-health graduate students, according to an email shared with The Post.
The first known deaths from the novel coronavirus were an 86-year-old woman and a 54-year-old man in King County on Feb. 26.
Two weeks later, the toll had reached 50. Four days after that, it topped 100. Then, 48 hours later, it had doubled.
Since March 21, the toll has increased by between 90 and 193 deaths per day, and on Wednesday, agencies reported nearly 250 fatalities, the most so far in the United States in a single day.
"We are at the beginning of the wave in most places in the United States," said Nahid Bhadelia, an infectious-diseases physician and medical director of the special pathogens unit at Boston University School of Medicine. "The worst is probably yet to come."
The United States now has the sixth-highest death toll in the world, behind Italy, Spain, China, Iran and France. In Italy, where more than a third of the world's virus-related deaths have occurred, 21 days passed from the first death to the 1,000th, recorded on March 13. From there, Italy's toll has climbed faster. Last weekend, it recorded 793 fatalities in a single day, the deadliest day of the outbreak anywhere.
Leaders and health experts in the United States have pointed to Italy as an example of what could happen as the American health-care system becomes overwhelmed and under-resourced.
Most victims had underlying medical conditions, hindering their immune system's response to covid-19's assault on their cells. In New York City, which had reported 280 deaths as of Wednesday evening, more than anywhere else in the country, 96 percent of people had a preexisting illness, such as asthma, diabetes, lung disease or cancer.
Yet the virus can also strike down those who were otherwise healthy.
James Carriere, a prominent local attorney and 10th-generation Louisianian, was one of those people. The 80-year-old was healthy, exercising regularly and enjoying family dinners in classic New Orleans haunts when he fell ill.
He was admitted to the hospital and died in quarantine about a week later.
His son, Olivier Carriere, said goodbye to him on FaceTime.
"He enjoyed life; he was always doing something. Then, all of a sudden -" he said. "We're all in shock."
Epidemiologists caution against becoming alarmed by the deaths of older people with no known underlying conditions, or by the story of a 35-year-old, seemingly in the prime of life, who succumbs to the disease.
"But the comparable data that you should have is: What about all the 35-year-olds who didn't die?" Branas said. "Without that, these cases are merely anecdotal."
"Some people have so many preexisting conditions that they are so deeply at risk that when they get the disease, it is very difficult to prevent their death," Branas added. "That's why you don't want to rely solely on mortality data."
To understand the likely trajectory of a disease, and who is most vulnerable, scientists need to be able to examine complete data on who survived, in addition to who died of the disease.
The Post's data on the first 1,000 fatalities reveals trends that already have emerged in studies from other countries that have been battling the outbreak far longer. There's a silver lining to this, Bhadelia said: If the disease were exacting a worse toll in the United States than in countries already ravaged - if it also killed young people at a high rate - that would have been borne out in these numbers. So far, that has not been the case.
Dense urban centers, many of them in coastal states, have been hit hardest in the first two months of the outbreak, but it's only a matter of time before the coronavirus takes hold in rural areas, too. In some places, such as Albany, Georgia, where at least 12 had died as of Thursday, it's already happening. When it arrives elsewhere, it could have a crippling effect, especially in places where resources and health-care workers are already in short supply.
"It might take longer for covid-19 to make it into the rural communities, and they might not get as many cases there," Bhadelia said, "but the worrisome thing is, it might not take as many cases to overwhelm the health-care system in these areas."
Or, as Cuomo put it earlier this week, warning that his state is the canary in the coal mine: "We are your future."
- - -
The Washington Post's Lena H. Sun, Jennifer Jenkins and Julie Tate also contributed to this report.
This article was written by Reis Thebault, Abigail Hauslohner and Jacqueline Dupree, reporters for The Washington Post.
As a public service, we've opened this article to everyone regardless of subscription status.