Poynter delivers one of the worst fact-checks yet, bizarrely declaring falsehoods in the public reading covid risk by age data themselves
This Poynter fact-check (link above) on COVID-19 survival rates is one of the worst I've seen yet, and so representative of the problems with fact-checking and the media's gatekeeping of information on the virus.
They fact-check an Instagram post (that's been removed as of writing) that according to them claimed the survival rate for covid is "over 99% for most age groups," comparing that to the effectiveness of the vaccine, given as 94%. The Instagram post breaks down the risk by age as follows: 99.997% for 0–19 years, 99.98% for 20–49, 99.5% for 50–69, and 94.6% for 70+.
Comparing the effectiveness of the vaccine with the survival rate data seems like a non sequitur but it's the covid risk data itself that they take most issue with, and on the overall fatality rate, the subject is clearly right: the whole-population infection fatality rate has long been widely estimated to be a little under 1% (Nature, BBC), making for a survival rate of +99%.
Poynter has no reasonable objection to this but declares it an "incorrect use of data" anyway. They claim that the post "improperly used" CDC data that's for modeling pandemic scenarios, not for calculating the survival rate, though there's no reason to think the data is in any way inaccurate — the whole point of that modeling is to get reasonable estimates of fatalities based on the data we have.
They attempt a few arguments but they're all bad.
We found that survival rates are usually calculated over a longer period of time, because death data can sometimes lag for months behind new cases.
This argument is repeated further down the article. It's probably true that patients who spend an extended period of time in hospital before dying contribute to an undercounting of the death toll, but the huge number of undetected cases among the general population works much more the other way to make the calculated case fatality rates (CFR) overestimates. They then go on to uncritically share this CFR data which is far more misleading than the numbers provided by the subject:
The CDC data shows that most people who have died from COVID-19, about 79%, have been people ages 65 and older. People between 45 and 64 years old account for about 18% of COVID-19 deaths, and people under 45 years account for 2.8% of such deaths, according to the CDC’s data as of Aug. 6. (The CDC data isn’t broken down in the same age groups offered in the Instagram post.)
Then there's this embarrassing logic failure:
Etzioni said that it’s not useful to just look at the rate that people die, even if it’s low, because it doesn’t tell the whole story. “If more and more younger people are getting COVID, then the total number of young people who die is going to skyrocket,” Etzioni said.
The death rate is a constant though (ignoring the effect of new variants, etc.) and this comment adds nothing to the debate.
They follow that comment up with this generic argument that risk varies according to a number of factors, which is obviously true, though it doesn't say much.
Also, people should not use data on how many people have survived COVID-19 to predict their own chances of surviving infection, experts say. Someone’s chances of surviving COVID-19 can vary depending on their age, health, and vaccination status — national statistics don’t account for these factors.
Regular people very much can look at the risk by age data and make reasonably informed conclusions for their own health. The gatekeepers simply don't like that people are doing so without their "guidance."
The piece ends by boldly declaring that "experts say a person cannot determine their own chances at surviving COVID-19 by looking at national statistics" and that the claims in the post are "false." Complete nonsense.
This isn't some random website inverting reality and deliberately lying about the facts while accusing others of spreading misinformation, the Poynter Institute is a leading authority on journalism and the media. They're described as being a journalism school but are also known for research, functioning something like an (allegedly) independent watchdog on the media. They operate PolitiFact, one of the most prominent fact-checkers in US media.
Almost amusingly, the piece says that the Instagram post was flagged as Facebook's efforts to "combat false news and misinformation," of which Poynter is noted to be a partner. Well that's a serious problem.
