Trends and prediction in daily incidence of novel coronavirus infection in China, Hubei Province and Wuhan City: an application of Farr law

Xu, Jie; Cheng, Yajiao; Yuan, Xiaoling; Li, Wei V.; Zhang, Lanjing

BackgroundThe recent outbreak of novel coronavirus (2019-nCoV) has infected tens of thousands of patients in China. Studies have forecasted future trends of the incidence of 2019-nCoV infection, but appeared unsuccessful. Farrs law is a classic epidemiology theory/practice for predicting epidemics. Therefore, we used and validated a model based on Farrs law to predict the daily-incidence of 2019-nCoV infection in China and 2 regions of high-incidence. MethodsWe extracted the 2019-nCoV incidence data of China, Hubei Province and Wuhan City from websites of the Chinese and Hubei health commissions. A model based on Farrs law was developed using the data available on Feb. 8, 2020, and used to predict daily-incidence of 2019-nCoV infection in China, Hubei Province and Wuhan City afterward. ResultsWe observed 50,995 (37001 on or before Feb. 8) incident cases in China from January 16 to February 15, 2020. The daily-incidence has peaked in China, Hubei Providence and Wuhan City, but with different downward slopes. If no major changes occur, our model shows that the daily-incidence of 2019-nCoV will drop to single-digit by February 25 for China and Hubei Province, but by March 8 for Wuhan city. However, predicted 75% confidence intervals of daily-incidence in all 3 regions of interest had an upward trend. The predicted trends overall match the prospectively-collected data, confirming usefulness of these models. ConclusionsThis study shows the daily-incidence of 2019-nCoV in China, Hubei Province and Wuhan City has reached the peak and was decreasing. However, there is a possibility of upward trend.