Phase-adjusted estimation of the COVID-19 outbreak in South Korea under multi-source data and adjustment measures: a modelling study

Abstract

<abstract> <p>Based on the reported data from February 16, 2020 to March 9, 2020 in South Korea including confirmed cases, death cases and recovery cases, the control reproduction number was estimated respectively at different control measure phases using Markov chain Monte Carlo method and presented using the resulting posterior mean and 95% credible interval (C_rI). At the early phase from February 16 to February 24, we estimate the basic reproduction number <italic>R</italic>₀ of COVID-19 to be 4.79(95% C_rI 4.38 - 5.2). The estimated control reproduction number dropped rapidly to <italic>R</italic>_{<italic>c</italic>} ≈ 0.32(95% C_rI 0.19 - 0.47) at the second phase from February 25 to March 2 because of the voluntary lockdown measures. At the third phase from March 3 to March 9, we estimate <italic>R</italic>_{<italic>c</italic>} to be 0.27 (95% C_rI 0.14 - 0.42). We predict that the final size of the COVID-19 outbreak in South Korea is 9661 (95% C_rI 8660 - 11100) and the whole epidemic will be over by late April. It is found that reducing contact rate and enhancing the testing speed will have the impact on the peak value and the peak time.</p> </abstract>