1 of 6:
Revisiting the original Imperial College 1st wave scenarios:
Firstly, 280 critical care beds/100k works out at 188k beds at the peak, which corresponds to 2.4m deaths, not 500k.
The 500k was later reduced to 350k, which would have given 39 critical care beds / 100k.
2 of 6:
Secondly, notice the odd order of interventions and the curiously equal benefit of each. This is normally an indication of a thumb-suck exercise.
Surely, it would be sensible to start with the least disruptive measure first, for example, case isolation.
3 of 6:
And then add on more disruptive measures such as measures focussing specifically on the vulnerable.
Limiting high risk gatherings and warning the public about high risk locations such as crowded and noisy indoor venues could then be added.
4 of 6:
The relative benefit of each measure is not known and purely shown for illustrative purposes.
It should however become clear why countries with much lighter interventions were not necessarily much worse off than the UK.
Low cost measures might get you 80% there.
5 of 6:
The highest cost measures should always be added last when producing conditional scenarios using mathematical modelling.
It is quite possible that the whole world overestimated the benefit and underestimated the cost of "stay at home orders".
6 of 6:
Sweden, which followed more sensible mitigation measures, while keeping most businesses open also came in well below original predictions.
Some more details on the calcs for the UK:
The actual numbers for the 1st wave were 41,000 deaths and 3,100 critical care beds i.e. 4.6 per 100,000.
The 350k deaths comes from this paper:
thelancet.com/journals/lanmi…
350k deaths would have resulted in 39 cc beds/100k.
Share this Scrolly Tale with your friends.
A Scrolly Tale is a new way to read Twitter threads with a more visually immersive experience.
Discover more beautiful Scrolly Tales like this.