ObjectiveTo analyze factors related to the suspected allergic reaction of elememe emulsion injection based on hospital information system. MethodData on cases that used elememe emulsion injection were collected from the information systems of 60 first-class hospitals nationwide. The nested case-control design method was adopted. Finally， 30 cases were included in the suspected allergy group and 120 cases in the control group. SAS 9.3 was employed for descriptive analysis of the gender， age， occupation， admission route， conditions of patients at the admission， and the diagnosis with frequency and percentage. The factors affecting the occurrence of suspected allergic reaction were analyzed by conventional logistic regression and propensity score weighted logistic regression. In the case that the number of independent variables was larger than the sample number， MCP （minimax concave penalty） was used to screen the key variables and the conditions of patients at admission， conditions of patients during hospitalization， hospital stay， diagnostic information， and medication information were compared between two groups. ResultThe male-to-female ratio was about 2∶1 in both groups and most of the patients were 46-65 years old. Patients in the control group were mainly "professional and technical personnel"， and the majority in the suspected allergy group were "business and service personnel" and "clerks and related personnel". They were mainly admitted at the outpatient and conditions of patients were average at the admission. Compared with the control group， suspected allergy group showed severe conditions during the hospitalization， short average hospital stay， large proportion with intravenous infusion， and low cure rate and effective rate. The results of logistic regression analysis showed no statistical difference in conditions of patients at admission， hospital stay， combined diseases， medicine dosage， and treatment course. ConclusionThe suspected allergic reaction of elememe emulsion injection mainly occurs in the first administration with rapid onset even with the dose lower than the commonly used one. The occurrence is related to the intravenous infusion and the severe conditions of patients during hospitalization and has nothing to do with the conditions of patients at admission， hospital stay， treatment course， use of other medicines， and diagnostic information. In summary， it is mainly related to the constitution and immune status of patients.