Risk map of the novel coronavirus (2019-nCoV) in China: proportionate control is needed

Background China is running a national level antivirus campaign against the novel coronavirus (2019-nCoV). Strict control measures are being enforced in either the populated areas and remote regions. While the virus is closed to be under control, tremendous economic loss has been caused. Methods and findings We assessed the pandemic risk of 2019-nCoV for all cities/regions in China using the random forest algorithm, taking into account the effect of five factors: the accumulative and increased numbers of confirmed cases, total population, population density, and GDP. We defined four levels of the risk, corresponding to the four response levels to public health emergencies in China. The classification system has good consistency among cities in China, as the error rate of the confusion matrix is 1.58%. Conclusions The pandemic risk of 2019-nCoV is dramatically different among the 442 cities/regions. We recommend to adopt proportionate control policy according to the risk level to reduce unnecessary economic loss.