As the development of Electronic Medical Records in hospitals, more and more "real world data" became available for clinical researches, especially for the clinical effectiveness evaluation. Besides traditional statistical methods, more Machine Learning methods are also used to analyze the data. In this study, one high value medical consumable -gel, which is used in large quantities in cleaning surgical incision, is analyzed using both a traditional statistical method and a Machine Learning method to evaluate its clinical applicability, efficacy and safety. The Electronic Medical Records for three years are collected including patient gender, age, quantity of gel usages, surgical incision grades, preoperative diagnosis, and information about postoperative recoveries and so on. Through the two analysis methods, the difference among incision healing, antibiotic use, and the postoperative hospital days after using gel or common saline in the wound are analyzed. The results show that the two methods can reveal different aspects in clinical evaluation. The decision tree classification can provide valuable suggestions for the reasonable use of the gel and making reasonable medical policy with the applicable conditions of gels.