Jump in U.S. Covid-19 Death Tally Expected as CDC Changes Counting Practices

Medical workers load a body into an ambulance at Andover Subacute and Rehabilitation Center on April 16, 2020 in Andover, New Jersey.
Medical workers load a body into an ambulance at Andover Subacute and Rehabilitation Center on April 16, 2020 in Andover, New Jersey.
Photo: Eduardo Munoz Alvarez (Getty Images)

The Centers for Disease Control and Prevention has changed how it will officially count the number of people sickened and killed by covid-19 in the U.S. This week, the agency announced it will start including in its tally both confirmed and probable cases and deaths attributed to the novel coronavirus. The decision was made on the recommendation of public health experts, according to the CDC.


In early April, the Council of State and Territorial Epidemiologists (CSTE), a nonprofit organization that represents epidemiologists in the field, issued an interim position statement on how covid-19 cases and deaths should be defined. They argued that the CDC and all health agencies in the U.S. should collect and present data on both confirmed and probable cases of the viral disease. They also laid out criteria for how cases should be classified as either confirmed or probable.

Any confirmed case or death would require a positive test for the genetic RNA presence of SARS-CoV-2, the coronavirus that causes covid-19. Someone with a probable case, on the other hand, may not have gotten this test but would still have at least two symptoms commonly reported with covid-19, such as fever, chills, dry cough, or a loss of smell and taste. In addition to symptoms, though, there would need to be at least one other form of evidence.


A probable case could have had close contact with someone else confirmed or suspected to have covid-19, recently traveled somewhere with a sustained outbreak, or be part of a high-risk group for catching it, such as health care workers. They could have also gotten other lab results that suggest an infection, such as a positive test for antibodies to the virus. Having a suggestive lab test and evidence of exposure would also be enough for a probable diagnosis.

For a death to be considered probable, meanwhile, the person’s death certificate (signed by a coroner or other medical practitioner) would need to list covid-19 or the coronavirus as “a cause of death or a significant condition contributing to death.” A confirmed covid-19 death would include the above as well as a positive RNA test result.

The CDC now states on its website that both confirmed and probable cases and deaths will be included in its provisional count, abiding to the guidelines set by the CSTE. With the CSTE’s blessing, the CDC also officially classified covid-19 as a nationally notifiable disease, which requires all state and local health agencies to provide any relevant case data to the CDC. As of April 14 2020, according to the CDC, there have been 605,390 cases and over 24,000 deaths attributed to covid-19 in the U.S.


Some people, largely on the political right, have accused governments of overcounting deaths caused by covid-19. But reporting has shown a consistent pattern of the opposite happening, with deaths likely due to the virus being undercounted in the U.S. and elsewhere. Cities across the country have reported spikes of people dying at home or in long-term care facilities in recent weeks, many of whom had never been tested for covid-19.

Ideally, these changes to the case and death criteria should make it easier to track the size and spread of local outbreaks. New York, for instance, revised its death toll earlier this week. It now has over 10,000 reported deaths, with nearly 4,000 considered probable (President Trump, who denied the severity of the U.S. covid-19 outbreak for months, quickly questioned the veracity of these new numbers).


Even these revised counts will probably still underrepresent fatalities, though. It’s not clear, for instance, just how far back in time health agencies and the CDC will go to count and add probable covid-19 deaths to the toll. And it’s not clear yet if and when all states will adhere to the CSTE guidelines, and how this data will be presented to the public (some third party trackers, such as the one created by Johns Hopkins University, now seem to include the revised death toll from New York).

It remains likely that it will take a long time for us to know with any certainty just how many people will have died of covid-19 by the pandemic’s end.


Science writer at Gizmodo and pug aficionado elsewhere

Share This Story

Get our newsletter


Manic Otti

“Some people, largely on the political right, have accused governments of overcounting deaths caused by covid-19"

Where do they get this shit? Is there a conservative think tank who’s sole purpose is to come up with bad ideas to disseminate to conservative nut job leaders and radio talk show hosts?