Cicali B, Michaud V, Knowlton C, Turgeon J.
Adverse drug events (ADE) are a major public health issue, and identifying which patients in a large population require targeted interventions can be quite difficult. Computational tools developed with clinical and pharmacological data can be highly valuable for identifying those patients. This project presents the application of a novel risk stratification tool that utilizes only medical claims data to identify members at high risk of ADEs in 2,528 members from a self-funded employer population. Algorithms were designed to score five different risk factors to personalize the patient’s risk for quick mitigation via health care professional interventions. In total, 15,911 medications were considered in the analysis, indicating an average of five medications per member (ranging from one to 48 medications per member). In total, the tool was able to identify 324 members (12.8%) considered at high risk for ADEs. Furthermore, 61 members (2.4%) considered at the highest risk for ADEs were identified by isolating those members who were in the high-risk groups for all five medication risk factors. In conclusion, our results indicate that a risk stratification tool based on medical claims cannot only quickly identify high-risk members but also can provide insights into how to intervene and prevent costly medical expenditures.Share this: