Here, as I showed in the update, are the reported deaths divided by reported cases for the world and US.
These are the crude case-fatality rates. They’re “cases” for a whole host of reasons which aren’t necessarily synonymous with traditional definitions of cases. That usually means symptomatic, or requiring treatment, or even showing up to the hospital. Here, it includes sporadic testing of even non-symptomatic testing.
Like I said, the real numbers of interest are the raw-fatality rate, dead/all people, infection-fatality rate, dead/infected, and case-fatality rate, dead/cases.
The virus is not killing 5%+ of the people it infects! Nor is it even killing 5%+ of cases. The crude CFR is exaggerates, by a large amount, the actual CFR, which nobody knows.
The only way to get the IFR and CFR is by sampling, i.e. intelligent measuring and testing and extrapolation. That has been done with previous diseases, and it’s started with coronadoom.
Depending on the source—like the CDC or consulting previous outbreaks—CFR estimates are from 0.3% to 1.3%, and IFR estimates are from 0.15% to 0.26%. These are reasonable numbers. The windows might not hold the eventual estimates, but they won’t be far off, either.
As of Wednesday night, there were 356,826 reported deaths and 5,781,642 reported cases worldwide. The gives a RFR of 356,826/5,781,642 = 6.2%. It is impossible this bug is killing 6.2% of those it infects, or even 6.2% of symptomatic people.
Yet we can use in the obvious way the estimated CFR and IFR, and the reported deaths, to estimate the actual number of infected, and actual symptomatic.
Here is the plot (which I finally figured out how to do; turns out you can use R’s base
ggplot2, as long as you start with
The number of estimated actual cases are anywhere from 25 to 120 million people. That is, about 0.3% to 1.6% of the world’s population had symptoms or were otherwise cases.
The number of estimated actual infections are anywhere from 140 to 240 million people. That is, about 1.8% to 3% of the world’s population are already infected.
Here’s the same for the US (which still dominates the world in reported deaths):
As of Wednesday night, and using our standard sources (which exaggerate death counts), there were 1,689,630 reported “cases” (positive tests) and 94,352 reported deaths. The crude CFR was 94,352/1,689,630 = 5.6%. Again, this bug is not killing 5.6% of those with symptoms. The RFR was 0.03%.
The number of estimated actual cases are anywhere from 8 to 30 million Americans. That is, about 2.4% to 9.1% of the US’s population had symptoms or were otherwise cases.
The number of estimated actual infections are anywhere from 37 to 62 million people. That is, about 11% to 19% of the US’s population are already infected.
If actual deaths are lower, then all these numbers will be too high.
In any case, it should be clear that there are an enormous number of already infected people, and a corresponding large number of un-tested-for actual cases. All we have to do to reveal who these people are is increase testing.
The graphs (which are driven by deaths, which we know are decreasing rapidly) show the infections and cases are slowing. Thus, test results will show what to a large extent is already there.
Yet we see in the media breathless reports of NEW CASES! Don’t trust me. Do this yourself. I typed into Herr Google “surge in new cases” and looked at the news feed (all stories within the last 24 hours of my checking):
- Seven Bay Area counties see surge in new COVID-19 cases
- With 114 new cases, Nepal records highest single day Covid-19 surge; national tally jumps to 886
- Northwoods health experts prepare for potential surge in COVID-19 cases
- Covid-19 cases surge in some states as Americans celebrate Memorial Day weekend
- Whitfield County makes national ranking with surge in COVID-19 cases, according to New York Times report
- Record jump in new coronavirus cases raising concerns of a surge in Austin
- Alabama sees largest increase in new COVID-19 cases so far this week
This was just on the first page of the search. There are many more to be had by substituting “spike” for “surge”. Nowhere did we find “mild expected increase”.
Now this could be the result of plain stupidity, and given our source is journalists this explanation has some weight to it. But for the same reason, it could also represent a purposeful attempt to highlight the only aspect of the terror they have left.
Whenever you see a report like this, show them these graphs.
To support this site and its wholly independent host using credit card or PayPal (in any amount) click here