 I'm Sierra Davis, and I'm a data scientist at the Children's Mercy Research Institute, and I'm presenting a poster titled, Comparing the Treatment Regimen of Newly Diagnosed Pediatric Leukemia Patients. We're going over the background, the methods that we use, the results, and our conclusions. Newly diagnosed pediatric leukemia patients with acute lymphoblastic leukemia are placed into risk groups based on the National Cancer Institute criteria and are treated using protocols defined by the Children's Archaeology Group, specific to each of those risk groups. Risk groups are not currently documented as a discrete diagnosis code, and sex notes are not accessible in a de-identified data warehouse. For this project, we developed processes to use a de-identified EHR to categorize newly diagnosed patients into risk groups and prepare their regimens across health systems. This project used the De-identified Health Facts for a day at a warehouse, which contains HIPAA-compliant data from participating institutions. Health systems without lab, medication, and procedure data were excluded. We identified a cohort of patients with ALL using ICD-9 and ICD-10 diagnosis codes and collected all available timestamped encounters for each patient. We developed an automated process to infer NCI risk groups and the date in which chemotherapy treatment was initiated based on the clinical information from a manual review. Data extraction, analysis, and visualizations were performed in our studio. We were able to identify 7,728 ALL patients with 191,108 encounters with available lab, medication, and procedure data. Manual analysis suggested that lumbar punctures were widely available within our data warehouse. We were able to identify 82% of our patients at standard risk and 18% of our patients as high risk, with a definable day one of chemotherapy. We developed timelines to compare treatments across health systems to evaluate if patients align with their recommended protocol by the children's oncology group. In conclusion, De-identified EHR data can be used to infer groups and impute the start of treatment for children with ALL. Visualizations provide deeper understanding of alignment between clinical behavior and recommended practices. Visualizations can be used for additional research to evaluate trajectories for patients with pediatric leukemia.