Prediction of H7N9 epidemic in China.

Background: In March 2013, human cases of infection with a novel A (H7N9) influenza virus emerged in China. The epidemic spread quickly and as of 6 May 2013, there were 129 confirmed cases. The purpose of this study was to analyze the epidemiology of the confirmed cases, determine the impacts of bird migration and temperature changes on the H7N9 epidemic, predict the future trends of the epidemic, explore the response patterns of the government and propose preventive suggestions.

Methods: The geographic, temporal and population distribution of all cases reported up to 6 May 2013 were described from available records. Risk assessment standard was established by analysing the temperature and relative humidity records during the period of extensive outbreak in three epidemic regions in eastern China, including Shanghai, Zhejiang and Jiangsu provinces. Risk assessment maps were created by combining the bird migration routes in eastern China with the monthly average temperatures from May 1993 to December 2012 nationwide.

Results: Among the confirmed cases, there were more men than women, and 50.4% were elderly adults (age >61 years). The major demographic groups were retirees and farmers. The temperature on the days of disease onset was concentrated in the range of 9°C-19°C; we defined 9°C-19°C as the high-risk temperature range, 0°C-9°C or 19°C-25°C as medium risk and <0°C or >25°C as low risk. The relative humidity on the days of disease onset ranged widely from 25% to 99%, but did not correlate with the incidence of infection. Based on the temperature analysis and the eastern bird migration routes, we predicted that after May, the high-risk region would move to the northeast and inland, while after September, it would move back to north China.

Conclusions: Temperature and bird migration strongly influence the spread of the H7N9 virus. In order to control the H7N9 epidemic effectively, Chinese authorities should strengthen the surveillance of migrating birds, increase poultry and environmental sampling, improve live poultry selling and husbandry patterns and move from a "passive response pattern" to an "active response pattern" in focused preventive measures.