Trying to make sense of the figures

I've been saying since the start of this health crisis that the numbers we are being shown are all just hooha or maybe Holohanha, a new term for all the mathematical masturbation the modellers and graphics studios are getting into.

The only meaningful numbers are:
  • Number of samples taken, daily and total
  • Number of samples tested, daily and total
  • Number of people re-tested, daily and total (one negative test does not mean never being at future risk)
  • Days between sample taken and test completed, best case, worst, mean
  • Days between test completed and results returned to subject best case, worst, mean
  • Number of positive tests, daily and total
  • Numbers admitted to hospital, daily and total
  • Numbers discharged from hospital, daily and total
  • Number of deaths, daily and total
Useful Data:
  • PPSN
  • Home eircode
  • Number in household
  • Work eircode
  • Employment classification (health-care worker [medic, nurse, physio, hospital porter, health-care assistant, medical admin, contractor to hospital/medical centre, medicall admin,etc]}
Forecasting peaks and all that stuff is just more Holohanha; we'll only know after we've passed the peak.

Any thoughts?
 
Last edited:
SPC100 said:
Iirc from other studies cfr is estimated to be more like .35%
Click to expand...

That is very interesting and it would be useful if you could find a source.

But my general point is that it is only when we get a sustained reduction in the number of Covid deaths, will we know that the virus stopped increasing 19 days earlier.

Agree on the attribution issues. However, we could assume that the error in attribution is constant.

Brendan
 
see disambiguation in this article between case fatality rate, infection fatality rate, crude mortality rate.

ourworldindata.org

Mortality Risk of COVID-19

Our interactive data visualizations that show the case fatality rate in each country are updated daily.
ourworldindata.org ourworldindata.org

If estimating number infected (in the past) from (today's) deaths the cfr is not the correct number to use.

I was researching at the time how the overall death rates had changed since covid, in an attempt to avoid attribution issues.
 
Last edited:
reason.com

Preliminary German Study Shows a COVID-19 Infection Fatality Rate of About 0.4 Percent

Preliminary results are out from a COVID-19 case cluster study in one of the regions worst hit by Germany's coronavirus…
reason.com reason.com

Over the last two weeks, German virologists tested nearly 80 percent of the population of Gangelt for antibodies that indicate whether they'd been infected by the coronavirus. Around 15 percent had been infected, allowing them to calculate a COVID-19 infection fatality rate of about 0.37 percent. The researchers also concluded that people who recover from the infection are immune to reinfection, at least for a while.
 
Hi SPC

What do those figures represent? Are they from the German study?

In calculating the death rate, would we not need to see a full cycle?

For example, out of everyone who got infected in February, they have either fully recovered now or died.

Brendan
 
They are current total irish deaths to date divided by the infection fatality rate (from the german study).

This gives an estimate of how many people in Ireland were infected.

Obviosuly we don't know what the IFR is in Ireland, if it turns out to be higher, less people would have been infected.
 
Ah, I see.

So 19 days ago, say 27 March, you have worked out that there were 110,000 people infected compared to the 2,121 confirmed cases figure published by the HSE.

Brendan
 
Bear in mind that Germany didn't have widespread BCG vaccinations, which are thought to have a signifcant impact on mortality rates.
 
You must log in or register to reply here.
Back
Top