The purpose of this study are to find useful knowledge through discovering relations and patterns of unknown facts from large data using data mining technique and to introduce a scheme of knowledge management concept in medical field. The application areas of data mining in medical fields include the medical utilization review analysis, disease pattern analysis, analysis related with health promotion and hospital management analysis. Among those areas, we selected the disease pattern analysis and studied on prediction of the diagnosis of hypertension patients. Three data mining techniques of the statistical analysis, decision tree analysis and C4.5 were performed on the health examination data from Korea Medical Insurance Corporation. From the experiments, the levels of importance of factors to hypertension were inferred and the specifications between hypertensive group and normotensive group was classified and identified. These results can be applied not only to the prediction of the diagnosis of hypertension patients but also to the medical decision support system for the management of hypertension. From now on, the data mining techniques that reproduce valuable information to help decision support will provide and be applied to various areas; clinical epidemiological study, useful information of health promotion project, health care policy support information. And the technique will also give the additional efficiency of national projects related health and the realization of scientific health social management resulting the much more national welfare service.