Prediction in therapeutic effectiveness research: Prolonged dose titration in warfarin patients and model transportability

摘要: Therapeutic effectiveness research relies heavily on prediction modeling, as improving therapeutic outcomes for individuals often requires being able to predict which patients are likely do poorly a given therapy. In this dissertation, we examine the specific case of starting warfarin therapy, many whom at higher risk bleeding and thrombotic events because they take long time determine their optimal dose. Additionally, general problem transportability clinical models whether that can be improved through sequential model updating. Specifically, conducted three projects with following goals: 1) To social, clinical, genetic factors associated maintenance dose in warfarin; 2) develop externally validate prolonged dose-titration these patients; 3) updating improve simulation study. Being experience titration could help clinicians decide use or less burdensome alternative oral anticoagulant. Furthermore, overall utility other greatly increased by strategies transportability, such Degree Type Dissertation Name Doctor Philosophy (PhD) Graduate Group Epidemiology & Biostatistics First Advisor Stephen Kimmel