Deborah Birx, the physician advising the White House’s coronavirus task force, gave voice to a real concern earlier this month. She told officials from the Centers for Disease Control that some of its numbers on mortality and case count could be inflated by up to 25 percent. “There is nothing from the CDC that I can trust,” the Washington Post reported Birx saying.
We now know more about what she was talking about, since Colorado has become the first state to publish two different numbers. One number is derived using a definition mandated by the CDC, which issued guidelines on March 24 specifying that “COVID-19 should be reported on the death certificate for all decedents where the disease caused or is assumed to have caused or contributed to death.” (Emphasis in original.)
The second is a narrower category called “deaths due to COVID-19,” which is limited to people who actually died of the disease. Counting in this way results in a decline of 23 percent in reported COVID-19 deaths. What’s going on here?
There have been many outlandish examples of people dying of things bearing not the slightest connection to COVID-19, but, having tested positive for the coronavirus, they were added to the official COVID-19 death tally anyway.
There was the Lehigh Valley, Penn., man who slipped and fell at home. The coroner said, according to LehighValleyLive.com: “The primary cause of the man’s death was a head injury from a fall at home, but that the virus was listed as a contributing factor to his death.”
In April a 37-year old California man “died as a result of a drug overdose while infected with COVID-19, a significant contributing condition, according to county spokeswoman Ashley Bautista, per the Ventura County Star.
Last week, there was the entry in the Cook County, Ill., medical-examiner ledger: “Complications of opiate (probably heroin) toxicity. Novel corona (COVID-19) virus, end stage renal disease, hypertension, ACCIDENT.”
Oddities, all, we were told. “Aberrations.” Insignificant to the overall count.
Except we know they were not.
Republican State Representative Mark Baisley accused the Colorado Department of Public Health and Environment (CDPHE) of fraudulently altering death certificates in cases in which coronavirus-positive individuals died but the medical examiner did not list COVID-19 on the death certificate. Baisley, in his request for an investigation, cited a letter from the Someren Glen nursing home to staff, residents, and their families saying that CDPHE had overruled physicians and added COVID-19 to seven death certificates.
CDPHE came under more fire when they added the death of a 35-year-old man who died of alcohol poisoning, who happened to be coronavirus-positive, to their official tally. Their defense was simple: They said they were just following the CDC definition. “We classify a death as confirmed when there was a case who had a positive SARS-CoV-2 (COVID-19) laboratory test and then died,” CDPHE said in statement.
Democratic governor Jared Polis could have circled the wagons. But to his enormous credit, he ordered the state to publish the two different numbers. He told Fox News Sunday that the new approach is “a better way to inspire confidence” in the data on the total number of virus victims.
The numbers differ considerably. At the time of Colorado’s announcement on Friday, the CDC-definition tally, used in CNN’s “dashboard” and all the other media reports, stood at 1,150 statewide. But only 878 of those, more than 23 percent less, are identified as deaths due to COVID-19.
This language appeared Friday on the state’s website, along with the new numbers:
Beginning May 15, the department began reporting the number of deaths in two ways:
The number of deaths among people with COVID-19. This represents the total number of deaths reported among people who have COVID-19, but COVID-19 may not have been the cause of death listed on the death certificate. This information is required by the CDC and is crucial for public health surveillance, as it provides more information about disease transmission and can help identify risk factors among all deaths across populations.
The number of deaths among people who died from COVID-19: This represents the total number of people whose death was attributed to COVID-19 as indicated on a death certificate. This number is determined by the CDC and is updated daily for dates through the previous Saturday.
There is no doubt this virus flashed through nursing homes with devastating effect. We have cataloged the death toll in these spreadsheets.
But if we don’t fix the definition, residents of nursing homes (which are now being tested wholesale for coronavirus) who have mild or asymptomatic infections will show up in the COVID-19death count when they die from any of the maladies from which people die. The median nursing-home stay before death is just five months. If this definitional problem isn’t fixed, the tallies will lose all contact with reality, with attendant public panic and additional economic hardship.
Colorado has taken the lead. Baisley lit the fire, and Polis stepped up to exercise genuine leadership. All governors, Republican and Democrat, should follow his lead and report both sets of numbers. And the CDC should fix its nonsensical definition — which didn’t even exist until March 24 of this year.
John Fund is National Review’s national-affairs reporter. Phil Kerpen is the president of the Committee to Unleash Prosperity and the co-editor of its Hotline.
More from National Review