Analysis Of The Single Disease Costs Based On OLAP And Data Minin

Rapid growth of high health care costs is a tough problem around the world. Based on the fundamental realities of the country, China put forward the single disease payment system. But there are several drawbacks in the implementation process of the single disease payment system. In order to identify the factors that affect the single disease cost and the specific impact model, which can provide the basis for improving the current payment system, the thesis summarizes the research status of the single disease payment and takes appendicitis for example to make an analysis on the single disease costs from multi-angle using data mining and on-line analytical processing (OLAP).First, we used DeepSee (a business intelligent software) to establish a multidimensional model which mixed the galaxies model and snowflakes model for the appendicitis patients. Then we carried out on the model various OLAP operations, such as slicing, dicing, rotation and drilling. Finally, we gave four examples to show the results of the modeling and analysis. The modeling results show that the proportion of western medicine fees in medical expenses is the largest. The results indicate that the hospital should strengthen supervision of medicine such as antibiotics to reduce drug costs. What’s more, the results also show that the social characteristics of patients, disease characteristics and socio-economic environment are all important factors that affect the patient’s medical expenses. These factors should all be taken into account in the analysis of the patient’s medical expenses.Further, this thesis used the data mining technology-decision tree to make an analysis for the appendicitis patient data in the hospital information systems in2005and2007. The results show that among all fields there are four factors affect the medical costs of appendicitis. These factors are the number of primary diagnosis, charge type, treatment days and patient’s condition. The model achieved high accuracy in both training data set and test data set. Several payment rules were extracted from the model. This thesis made a multi-angle analysis for the single disease costs and put forward some recommendations. It provides a reference for the graded payment system of the single disease.