Predicting the trend of the COVID-19 outbreak and timely grading the current risk level of epidemic based on moving average prediction limits / 上海交通大学学报（医学版）

ABSTRACT

Objective: To establish a practical data-driven method that helps predict the evolutionary trend of the coronavirus disease 2019 (COVID-19) epidemic, track and prejudge the current risk classification of the epidemic area, and provide a quantitative evidence for precision prevention and control strategies. Methods ¡¤ A moving average prediction limit (MAPL) method was established based on the moving average method. The previous severe acute respiratory syndrome (SARS) epidemic data was used to verify the practicability of the MAPL method for predicting epidemic trends and quantitative risk. By tracking the COVID-19 outbreak epidemic data publicly reported since January 16, 2020, the MAPL method was used for timely epidemic trend prediction and the risk classification. Results ¡¤ According to the MAPL analysis, the na-tional epidemic of COVID-19 peaked in early February 2020. After active prevention and control in early stages, the overall epidemic situation in the country showed a downward trend from mid-February to mid-March. Compared with Hubei Province, the number of new cases in non-Hubei region declined rapidly in mid-February, but then increased slightly. The analysis of imported cases since March showed that there was a medium to high level of epidemic import risk in the near future. It is recommended to take corresponding prevention and control measures to prevent the epidemic from spreading again. Conclusion ¡¤ The MAPL method can assist in judging the epidemic trend of emerging infectious diseases and predicting the risk levels in a timely manner. Each epidemic district may implement a differentiated precision prevention and control strategies according to the local classification of epidemic risk. Since March, attention should be paid to the prevention and control of imported risks.