I’m particularly interested in people sharing models / spreadsheets that they’re currently working through here. (Posting them as top level posts is also fine, but I thought it might be good to encourage more “thinking out loud” in quantitative ways, even while you’re still fleshing a model out and still have a lot of open questions about it)
Buck’s tentative Guesstimate model of Wei Dai’s “hospital crowding” catastrophic scenario. Many folks have already seen his comment, but I’m posting a link to it for completeness.
Daniel Filan’s Guesstimate model on whether he should stay home for work. Many folks have already seen his comment, but I’m posting a link to it for completeness.
When to cancel events due to Coronavirus? Calculations by Linch Zhang [1], I’ve put them into a Guesstimate with some slight changes and adaptions for Austria [2]
Under my assumptions, if 1 in 7700 people gets newly infected every day, it translates to an infection risk of 0.2% per encounter (range of 0.45% − 0.053%). Feedback welcome.
Unfortunately it doesn’t let you modify the assumptions about disease severity or number of undetected cases. It assumes that the majority of cases have been undetected (which seems questionable) and that 4.31% of cases are severe (which seems low even if the majority are undetected). It gives a case fatality rate of 0.97%, which doesn’t seem to depend on any of the other parameters.
In their baseline scenario (for a small Swiss city with good infection control) 0.26% of the population dies.
With no infection control this goes up to 0.76% of the population dying, with no change in the CFR.
If you also increase the length of a hospital stay from 10 days to 20 days, the total number of deaths actually decreases slightly because the spread is slower. So while the graph is a nice way to see how long hospitals will be overwhelmed in different scenarios, it doesn’t show you anything useful about how this affects outcomes. I would love to be able to add in some parameters for fatality rate for severe and critical cases with/without a hospital bed.
In their baseline scenario (for a small Swiss city with good infection control)
Sad laugh. I’m in Switzerland, we have exponential growth and there’s no infection control to speak of. They just told people with non-severe symptoms to not bother getting tested. Schools are open. Haven’t seen even one person wearing a mask.
The Medium article that Wei Dai cited in his comment links to an “open-source model”. I haven’t examined it closely, though I did notice that some of the formulas are weirdly constructed (e.g. using INDIRECT rather than absolute cell references) and that some of the assumed parameters are overly pessimistic (e.g. a 3.4% CFR).
I’m particularly interested in people sharing models / spreadsheets that they’re currently working through here. (Posting them as top level posts is also fine, but I thought it might be good to encourage more “thinking out loud” in quantitative ways, even while you’re still fleshing a model out and still have a lot of open questions about it)
Here’s my bay area hospital capacity model: https://www.getguesstimate.com/models/15278
Here’s a basic SIR model created by Metaculus user Isinlor. (I haven’t looked at it, so don’t interpret this comment as an endorsement.)
Buck’s tentative Guesstimate model of Wei Dai’s “hospital crowding” catastrophic scenario. Many folks have already seen his comment, but I’m posting a link to it for completeness.
Daniel Filan’s Guesstimate model on whether he should stay home for work. Many folks have already seen his comment, but I’m posting a link to it for completeness.
When to cancel events due to Coronavirus? Calculations by Linch Zhang [1], I’ve put them into a Guesstimate with some slight changes and adaptions for Austria [2]
[1] https://docs.google.com/document/d/1A0jcxj4n0BvNt_jMunHT5WSsAKFzuVJJyaaqcK9Z1HU/edit#
[2] https://www.getguesstimate.com/models/15367
Dating during Coronavirus: What’s the risk of going on a date with a random new person at the height of an outbreak? https://www.getguesstimate.com/models/15381
Under my assumptions, if 1 in 7700 people gets newly infected every day, it translates to an infection risk of 0.2% per encounter (range of 0.45% − 0.053%). Feedback welcome.
Here’s a tool to estimate how badly hospitals will be overfilled in your community.
http://scratch.neherlab.org/
It’s by Richard Neher and colleagues and an early stage tool. Might nevertheless be interesting to play around with.
Here’s the source and some explanations about the underlying model:
https://twitter.com/richardneher/status/1236980631789359104
Unfortunately it doesn’t let you modify the assumptions about disease severity or number of undetected cases. It assumes that the majority of cases have been undetected (which seems questionable) and that 4.31% of cases are severe (which seems low even if the majority are undetected). It gives a case fatality rate of 0.97%, which doesn’t seem to depend on any of the other parameters.
In their baseline scenario (for a small Swiss city with good infection control) 0.26% of the population dies.
With no infection control this goes up to 0.76% of the population dying, with no change in the CFR.
If you also increase the length of a hospital stay from 10 days to 20 days, the total number of deaths actually decreases slightly because the spread is slower. So while the graph is a nice way to see how long hospitals will be overwhelmed in different scenarios, it doesn’t show you anything useful about how this affects outcomes. I would love to be able to add in some parameters for fatality rate for severe and critical cases with/without a hospital bed.
Sad laugh. I’m in Switzerland, we have exponential growth and there’s no infection control to speak of. They just told people with non-severe symptoms to not bother getting tested. Schools are open. Haven’t seen even one person wearing a mask.
Another basic SIR model, which considers impacts on hospital capacity (and resulting deaths) from infection controls of various degrees.
Coronavirus automatic tracking and population modeling v.2
The Medium article that Wei Dai cited in his comment links to an “open-source model”. I haven’t examined it closely, though I did notice that some of the formulas are weirdly constructed (e.g. using INDIRECT rather than absolute cell references) and that some of the assumed parameters are overly pessimistic (e.g. a 3.4% CFR).