Mining Rules for Risk Factors on Blood Stream Infection in Hospital Information System

Since a large volume of clnical data has been stored in a hospital, data mining is coming to be important for decision making and risk management in a hospital. In this paper, C4.5 and statistical techniques were applied to extract patterns from our hospital clinical microbiology databases and to analyze the effects of lactobacillus therapy and the background risk factors on blood stream infection in patients. The significant "If-then rules" were extracted from a decision tree between bacteria detection on blood samples and patients' treatments, such as lactobacillus therapy, antibiotics, and various catheters.. Then, chi-square test, odds ratio and logistic regression were applied in order to analyze the effect of lactobacillus therapy on bacteria detection. Experimental results suggest that lactobacillus therapy may be effective in reducing the risk of blood stream infection. Especially, rule induction (C4.5) is useful for extracting background risk factors of blood stream infection from our clinical database.