Update of intuitively understandable plot on current #COVID19 situation in DE. Plot of #livesatrisk calculated from age distribution and age-specific risk. Lives at risk for the next two weeks still rising due to infections in the most vulnerable groups. This is bad news. (1/6) Image
Inclusion of overall infections on sec axis (shifted by 2 wks) illustrates decoupling between infections and #livesatrisk. While infections are stable for 4 wks now the mortality still rises! Pre-summer the development was in parallel, indicating something is different now (2/6). Image
This needs to be investigated. Because the ratio of actual to predicted deaths has risen fast to 53%, the change in testing strategy obviously plays a role, leading to more undetected infections. The policy change to leave nursing homes open may also play a significant part (3/6) Image
1w ago a sad new record was correctly predicted, but underestimating real dynamics (1750 deaths expected for CW48 vs 2101 found). If this holds we expect another record 2400 and 2560 #COVID19 fatalities in DE in this and next week. I hope it is not another underestimation (4/6).
For completeness, the plot without correction factor, i.e. #livesatrisk based on pre-summer age-specific deaths. This can be viewed as the "ceiling" for mortality prediction. I expect the ratio of real to predicted deaths will not exceed ~70% due to improvements in therapy. (5/6) Image
The continueing good correlation further supports the use of #livesatrisk plots to allow focus on the most relevant parameter. Ratio of actual deaths to lives remained stable enough for robust results but was rising strongly most likely due to changed test strategy (6/6).
Added note: As always, discussion, doubts and questions very welcome! Age-dependent infection data obtained from RKI. Deaths from ECDC. Age specific mortality calculated using ECDC data from CW10-CW30 and RKI daily report data.
Now #livesatrisk also available for individual federal states of Germany.
Please find a website with #livesatrisk data for all German cities and regions under andman.eu. Description in the linked tweet. https://t.co/A8soqhB65p

• • •

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

Keep Current with Carl Jakobs

Carl Jakobs 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 @JakobsCarl

6 Dec
Das Alter der Person, die sich mit #COVID19 infiziert ist entscheidend für das Sterberisiko. Die Infektionszahlen bilden dies nicht ab, der Parameter #livesatrisk schon. Ich stelle ihn für die letzten 5 KW für alle Kreise in DE auf andman.eu zur Verfügung. (1/x)
Die Analyse funktioniert durch Multiplikation der Infektionszahlen für jede Alterskohorte mit ihrem altersspezifischen Sterberisiko nach #COVID19 Infektion. Dazu verwende ich die Zahlen aus Deutschland von KW10-KW30 wie in verlinktem Tweet gezeigt. (2/x)
Die letzte Analyse für Deutschland insgesamt findet sich im verlinkten tweet. Hier ist deutlich zu erkennen, daß die #livesatrisk im Gegensatz zu den Infektionen weiter ansteigen (vgl frühere tweets). Die stetig steigende Mortalität bestätigt dies. (3/x)
Read 9 tweets
4 Dec
Ein genauerer Blick auf die durch #COVID19-Infektionen riskierten Leben #livesatrisk nach Bundesland. Hier absoluten Zahlen. Damit lässt sich die Mortalität über diese und nächste Woche vorhersagen. Zudem erlaubt es weitere spannende Einsichten. Städte demnächst. (1/6) Image
Einige Bundesländer und die Stadtstaaten machen es besser als Andere. Zuvorderst Hamburg, aber auch Bremen, Schleswig-Holstein und das Saarland und Niedersachsen scheinen das Wachstum an riskierten Leben aufgehalten zu haben (hier die #livesatrisk pro 1M Einwohner). (2/6) Image
Warum ist Hamburg so gut? Der Schulterschluss von Bürgermeister @TschenPe und Experten wie @ChanasitJonas, Addo, Kluge und @degampraesident scheint effektiv. Darüber wüsste ich gern mehr. Wer weiss etwas dazu, was macht Hamburg anders als der Rest? (3/6) aerztezeitung.de/Nachrichten/Po…
Read 11 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 Become our Patreon

Thank you for your support!

Follow Us on Twitter!