The best available modeling of COVID-19 has many shortcomings. This is a game that will be fought with data due to the epidemiological parameters of the virus.
I believe we need better mathematical models, deeper personal data, and crucially a better understanding of the sensitivity of the modeling. This is to inform policy around confinement measures, when and where to strengthen it and relax it.
The mathematical modeling is starting to be discussed here. The author of the blog post has done the very very valuable work of digging through the Supplementary Information in all those papers.
Hopefully we can:
- get a better understanding of the sensitivity of those models, which means we will know better what data to crowdsource
- try to develop some better mathematical modeling, based on informed data. There are some leads here.
As a consequence we should know better what apps or services to implement in the global hackathon currently in the making, particularly for Switzerland.