Patient-Type Bayes-Adaptive Treatment Plans

Treatment decisions that explicitly consider patient heterogeneity can lower the cost of care and improve outcomes by providing right for at time. “Patient-Type Bayes-Adaptive Plans” analyzes problem designing ongoing treatment plans a population with in disease progression response to medical interventions. The authors create model learns type monitoring health over time updates patient's plan according information gathered. formulate as multivariate state space partially observable Markov decision process (POMDP). They provide structural properties optimal policy develop several approximate policies heuristics solve problem. As case study, they data-driven decision-analytic study timing vascular access surgery patients progressive chronic kidney disease. further insights sharpen existing guidelines.