By studying the distribution rules and characteristics of research variables in statistical information on first pages of medical records, the paper discusses the contribution of each research variable to solving the problem whether medical records should be included in clinical pathways. It introduces data sources and the preprocessing process and explains the selection of research method and specific process of data analysis and mining. Then, it constructs the Logstic regression equation of research variables and finds out features of medical records which have a high probability to be included in clinical pathways.