This study investigates an analytics approach to improving clinical outcomes, which is to utilize EHR data to identify and exploit variety of successful treatments that physicians deliver for different types of patients. A simulated environment was used to develop and test the proposed approach that assigns patients to physicians for better treatment of type-2 diabetes in a clinic setting. Results from this study demonstrate a 30% increase in patients being successfully brought to A1c goal, a 30-fold decrease in treatment errors, and a 12% decrease in outpatient costs to treat patients compared to traditional treatment procedures in diabetes clinic. These techniques can be integrated with EHR systems, thus helping hospitals meet Stage 3 meaningful-use requirements for improving patient outcomes.