An epidemiological forecast model and software assessing interventions on COVID-19 epidemic in China

Peter X Song; Lili Wang; Yiwang Zhou; Jie He; Bin Zhu; Fei Wang; Lu Tang; Marisa Eisenberg

It is worth pointing out that the estimates of the reproduction numbers obtained from the epidemiological models with time-varying transmission or quarantine rates appear larger than those obtained from the basic model with no quarantine. This is no surprising as our new models explicitly incorporate interventions, so that the estimated R 0 is an adjusted number with the influence of interventions be removed. In contrast, the basic model with no use of the quarantine modifier implicitly integrates the effect of interventions into the transmission rate β, and consequently the estimated R 0 is reduced due to the contribution from interventions. Our analyses suggest that reproduction numbers R 0 of COVID-19 without public health interventions would be around 4-5 within Hubei and around 3-3.5 outside Hubei with relatively big credible intervals. These findings are in agreement with findings from [17] . As pointed out above, the size of credible interval may be reduced with more accessible data of COVID-19, which permit users to specify smaller variances in the prior distributions given in section 3.1.