Retrospective prediction of the epidemic trend of COVID-19 in Wuhan at four phases

The coronavirus disease 2019 (COVID-19) outbreak caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) began in December 2019 and was basically under control in April 2020 in Wuhan. To explore the impact of intervention measures on the COVID-19 epidemic, we established susceptible-exposed-infectious-recovered (SEIR) models to predict the epidemic characteristics of COVID-19 at four different phases (beginning, outbreak, recession, and plateau) from January 1st to March 30th, 2020. We found that the infection rate rapidly grew up to 0.3647 at Phase II from 0.1100 at Phase I and went down to 0.0600 and 0.0006 at Phase III and IV, respectively. The reproduction numbers of COVID-19 were 10.7843, 13.8144, 1.4815, and 0.0137 at Phase I, II, III, and IV, respectively. These results suggest that intensive interventions, including compulsory home isolation and rapid improvement of medical resources, can effectively reduce the COVID-19 transmission. Furthermore, the predicted COVID-19 epidemic trend by our models was close to the actual epidemic trend in Wuhan. Our phase-based SEIR models demonstrate that intensive intervention measures can effectively control COVID-19 spread even without specific medicines and vaccines against this disease.

Automated summary

This study used SEIR models to predict the epidemic characteristics of COVID-19 in Wuhan at four different phases, showing that intensive intervention measures can effectively reduce the infection rate and reproduction numbers of the virus, and the predicted trend by the models closely matched the actual epidemic trend.