Context-Aware Optimization of Follow-Up Intervals for Type 2 Diabetes Care Using Markov Decision Processes
Researchers have developed a Contextual Markov Decision Process (CMDP) model to optimize follow-up intervals for Type 2 Diabetes (T2D) patients, moving beyond the American Diabetes Association's fixed guidelines. By analyzing electronic health records from over 22,000 patients, the model identified two distinct risk subpopulations. The CMDP-derived policies recommend adaptive follow-up schedules, suggesting intervals from 1 month for unmeasured labs to 6-12 months for sustained glycemic control, with shorter intervals for higher-risk patients. This approach demonstrated a significant reduction in expected cumulative cost compared to fixed-interval benchmarks. AI
IMPACT This research demonstrates how AI can personalize chronic disease management, potentially leading to more efficient and cost-effective healthcare delivery.