I made some more calculations.

This time comparing the death rates between #Delta and #Omicron in South Africa.

TL;DR: Death rates so far are fully consistent with a wave of similar (or even higher) severity as Delta.

(1/x)
Studies suggest that patients tend to pass away about 18 days after hospitalization.

thelancet.com/journals/lance…

Since South Africa publishes data on a weekly basis, a 3-week time lag between positive tests and fatalities seems most appropriate.

(2/x)
South Africa uses May 9 as the starting point for the #Delta wave (April 11 seems more appropriate, but let's go with their date).

(3/x)
If we calculate the ratio of cases and deaths during the ascending phase of the #Delta wave with a 3-week lag, we can calculate the ratio as follows:

393 deaths (in epiweek 22) / 19,935 cases (in epiweek 19) = 1.9%

nicd.ac.za/wp-content/upl…

nicd.ac.za/wp-content/upl…

(4/x)
Going on like that we get the following numbers:

EW 23/20: 566 / 24,212 = 2.3%
EW 24/21: 704 / 29773 = 2.4%
EW 25/22: 1032 / 36102 = 2.9%
EW 26/23: 1358 / 59445 = 2.3%
EW 27/24: 1498 / 87937 = 1.7%
EW 28/25: 1655 / 118488 = 1.4%
EW 29/26: 1554 / 146543 = 1.1%

(5/x)
Now let's compare this with the burgeoning #Omicron wave. South Africa uses November 7 as a starting point of the wave (which makes sense).

We have fairly complete data for deaths in epiweeks 48 and 49 and incomplete data for epiweek 50.

(6/x)
The same ratio for the #Omicron wave with the same 3-week time lag is:.

Epiweek 48/45: 104 / 2263 = 4.6%
Epiweek 49/46: 221 / 4741 = 4.7%

With an appropriate time lag, the death rate of #Omicron is HIGHER than in the #Delta wave.

(7/x)
For Epiweek 50, fatality numbers are incomplete and currently stand at 287, will likely be revised upward.

Epiweek 50/47: 287 / 18,775 = 1.5%

Even with this INCOMPLETE number, the death rate is still in line with the #Delta wave when Delta grew most strongly (EWs 24-26).

(8/x)
In summary, the supposedly low death rates so far are fully consistent with a new wave that will be at least as severe as the #Delta wave.

South Africa will hit Delta-level severe outcomes in about 3 weeks from now, and might conceivably go even higher.

(9/x)
I have made the same analysis for hospitalizations with exactly the same result. The data is completely consistent with a wave at least as severe as #Delta.

(10/10)
Please note that I have made an updated new thread on the death rates because I had to correct some numbers.

These updates DO NOT change any of the conclusions of this thread.
I posted a correction of this thread. The conclusions have NOT changed, but a data lapse has been corrected.

Please find the updated thread below.

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Trump Critic

Trump Critic Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @TrumpCritic

19 Dec
South Africa: I have made some calculation about the ratio of hospitalizations to cases in the #Omicron and #Delta waves.

TL;DR: The ratio is dependent on the pace of the spread and Omicron is so far just a little lower than Delta.

(1/x)
I am using these data sets here:

Cases:
nicd.ac.za/wp-content/upl…

Hospitalizations Delta:
nicd.ac.za/wp-content/upl…

Hospitalizations Omicron:
nicd.ac.za/wp-content/upl…

(2/x)
South Africa's presentation I commented on the last few days uses May 9 as a starting point for the #Delta wave.

On May 9, South Africa reported 1,790 new cases (7-day average).

That means the wave began with comparatively low growth by absolute numbers.
Read 19 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

:(