Thread by @jukka235

Denmark/Omicron 18 Dec 2021: first look at people in hospital data. Looks to me that people who are in hospital have tested positive 6-7 days earlier.
1/x

Also it would appear that the amount of people in hospital is about 1.9-2.2 percent of people who tested positive 6-7 days earlier. That level is 79-83 percent lower than the overall level of people in hospital in Denmark before the Omicron wave (10.3-10.8%).
2/x

How each point in this graph has been calculated:
- reliable data on new cases exists only up to 13 Dec
- first example (red circle): comparing people in hospital 15-17 Dec with new cases 11-13 Dec (=4 days earlier)
3/x

1) people in hospital: sum of last 3 days = 52
2) new Omicron cases 4 days earlier = 5505
3) 52/5505 = 0.94 percent (data point for 17 Dec)
4/x

Example 2: calculating all the data points on the 7 day line.
5/x

Here are all the data points calculated on the 7 day line. Example: data point for 17 Dec = 52/2357 = 2.21 percent.
6/x

Here is a cumulative version of the above graph. There appears to be initial convergence towards the 6-7 day range (which would mean that people are at hospital an average of 6-7 days after having tested positive with Omicron).
7/x

Here is a quick calculation of length of hospital stay
- hospital stay appears roughly 0.85 days
- therefore: admissions to be measured at the 5-6 day point
8/x

Therefore Omicron admissions in Denmark, measured at the 5-6 day mark, are estimated at 1.7-2.1 percent of people who tested positive 5-6 days earlier.
9/x

Summary estimates:
- roughly 1.7-2.1 percent of people, who have tested positive 5-6 days earlier, are admitted to hospital, where they stay an average of 0.85 days
- roughly 1.9 - 2.2 percent of people who tested positive 6-7 days earlier are at hospital
10/x