https://www.medrxiv.org/content/10.1101/2020.05.02.20088120v1.full.pdf
- some similarity in model and general approach
- use fewer disease compartment distinctions, but include age and ??
- considered (separetely) PCR and serological testing
- focussed on biases in CFR and IFR due to choices about testing. Didn't look at Rt, r, peak timing etc.