OBJECTIVE： To explore and analyze signals of Xiyanping injection-induced urticaria from spontaneous reporting system database of Jiangxi province. METHODS： The continuous ADR report monitoring data were collected from spontaneous reporting system of Jiangxi province， and established database. Bayesian confidence propagation neural network（BCPNN） method was used to calculate information component， IC value of Xiyanping injection-induced urticaria. The year-to-year changes of IC value and its interval were analyzed. RESULTS： A total of 328 324 ADR reports were collected in Jiangxi province during 2004-2016. The IC value of Xiyanping injection-induced urticaria detected by BCPNN method was 1.10 （the lower limit of IC value was 0.65， and upper limit was 1.54）， i.e. there was signal and the IC value ranged from -0.87 to 1.65.  CONCLUSIONS： The results of BCPNN method suggest that urticaria is a warning signal of Xiyanping injection. The risk is increasing gradually， and prediction precision increase with the addition of report quantity.