Background: Few evidences were revealed about the need of different approach in rapid response system (RRS) for specific patients’ department. Methods: We reviewed patients who admitted to 1300-bed tertiary, academic hospital and detected by the RRS from October 2012 to February 2014. We compared the triggers, interventions, and clinical outcomes between patients admitted to the medical and non-medical services. Results: Overall, 21,415 alert lists were generated from 11,271 patients. The RRS was activated in 460 patients, with the incidence of activation being almost three times higher in the non-medical service (74.3%) than in the medical service (25.7%). The triggers for RRS activation significantly differed between the medical and non-medical groups(p = 0.001), which were in order as abnormal respiratory rate (27.1%), abnormal blood gas analysis (23.7%), and low blood pressure (22%) in the medical group, as compared with low blood pressure (30.1%), low oxygen saturation (19.6%), and abnormal heart rate (16.7%) in the non-medical group. In the case of interventions performed following RRS activation, patients were more likely to be classified as do not resuscitate (DNR) or require intensive care unit admission in the medical group than the non-medical group (65.3% vs. 54.7%; p = 0.045). Difference of individual interventions was also statistically significant. Although the majority of patients were alive at discharge from the hospital in both groups, the number of alive patients was significantly greater in the non-medical group than the medical group(86.8% vs. 60.2%; p ＜ 0.001). Conclusion: We found that the triggers, interventions, and clinical outcomes associated with the RRS greatly differed between the medical and non-medical services. Further research is needed to evaluate the efficacy of a tailored approach to specific groups in the RRS.