Inferring COVID-19 spreading rates and potential change points for case number forecasts

Jonas Dehning; Johannes Zierenberg; Frank Paul Spitzner; Michael Wibral; Joao Pinheiro Neto; Michael Wilczek; Viola Priesemann

The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/2020.04.02.20050922 doi: medRxiv preprint presumably followed by the public in a timely fashion. Continuous changes, originating e.g. from increased awareness of the population can be accounted for by the discrete steps as well, within the precision of reported cases we have. For the spreading rates, we chose log-normal distributed priors as in the simple model. In particular, we chose for the initial spreading rate the same prior as in the simple model, λ 0 ∼ LogNormal(log(0.4), 0.5); after the first change point λ 1 ∼ LogNormal(log(0.2), 0.5), assuming the first intervention to reduce the spreading rate by 50% from our initial estimates (λ 0 ≈ 0.4) with a broad prior distribution; after the second change point λ 2 ∼ LogNormal(log(1/8), 0.2), assuming the second intervention to reduce the spreading rate to the level of the recovery rate, which would lead to a stationary number of new infections. This corresponds approximately to a reduction of λ at the change point by 50%; and after the third change point λ 3 ∼ LogNormal(log(1/16), 0.2), assuming the third intervention to reduce the spreading rate again by 50%. With that intervention, λ 3 is smaller than the recovery rate µ, causing a decrease in new case numbers and a saturation of the cumulative number of infections.