MetroCare Health System—Leveraging AI and social data for dynamic risk stratification

This teaching case explores MetroCare Health System’s innovative use of artificial intelligence (AI) to improve risk stratification for patients with type 2 diabetes (T2D). Facing substantial operational and financial pressures due to rising chronic disease management costs, MetroCare sought to enhance its traditional risk models, which relied solely on structured clinical data and often misclassified intermediate-risk patients. Leveraging natural language processing (NLP) and machine learning, MetroCare integrated previously untapped social and behavioral determinants from clinicians’ unstructured electronic health record notes into a new predictive model. This AI-driven approach significantly improved accuracy (AUC increased from 0.834 to 0.860), reduced unnecessary clinical interventions by approximately 30%, and generated estimated annual cost savings of about $1.7 million per 1000 patients. The case highlights critical implementation considerations, including threshold tuning to balance intervention precision against potential missed risks, workflow integration challenges, clinician acceptance, and ethical issues surrounding patient privacy and fairness. It illustrates the strategic alignment of innovative IT solutions with organizational goals, offering graduate and executive learners a realistic context to discuss how emerging technologies can transform healthcare operations. Ultimately, students are tasked with determining how best to operationalize this data-driven solution to improve clinical outcomes, optimize resource allocation, and advance MetroCare’s mission in a competitive, value-based healthcare environment.