As lots of research has discussed, clinical pathways provide an effective way for improving the efficiency of hospitals. However, how to find useful clinical pathways conveniently is a problem. With the rapid development of networking, data storage and the data collections capacity, the hospitals have accumulated a large number of clinical data. In this paper, we characterize the problem of mining clinical pathways as a sequential patterns mining problem. We propose the concept of prefix of CP and integrate the prefix set into our algorithm CPM-PC:Clinical Pathways Mining with Prefix Constraints. The algorithm is more suitable for mining clinical pathways and will not search the sequences that has no medical significance. And the method has been applied to a real world data set to find clinical pathways and performs well.