# Confirmed cases vs actual cases

#### Pwave

The number of cases reported are for confirmed cases. Is there a published number of likely actual cases?

Here is my estimation method.

If one observes the death total for a given day, then assume that those people probably contracted the virus 14-21 days prior, one could additionally assume that the inverse of the deaths rate number of infections produced that total of dead. If we use that hypothesis, a WHO expected deaths rate of 3.6%, a 22 Mar total dead of 413, one calculates that 11,422 infections generated those deaths.

If we look back 14 days from 22 Mar, we find 106 new cases were identified. That is an undercounting of cases by a factor of ~100. Assuming that same undercounting in still occurring, we have 1,000,000 new cases every day.

Choosing a lower death rate or longer average time to death only makes the numbers worse.

Feel free to dissect my method

(Sorry for the cross post in “Any coronavirus questions...”)

Matthew in Ohio

#### jpishgar

It's a decent method - though the undercounting tends to vary by region and country based on the availability of kits.

I'm seeing that the lethality appears to vary based on government type and time to issue stay-at-home orders. What sources are you deriving data points from? WHO overall, or by nation?

Matthew in Ohio

Replies
23
Views
68K
Replies
0
Views
802
Replies
0
Views
1K
Replies
2
Views
3K
Replies
0
Views
1K