My research is aimed at developing new methodological tools that will address a variety of issues essential to answering important patient-centered clinical questions such as "How can we optimize the sequence of specific treatments for specific patients, given their personal characteristics, conditions and preferences?" Clinical and research databases, collectively called electronic health records, may provide unprecedented opportunities to learn about treatment individualization, however many methodological challenges exist before their full power may be harnessed.

My overarching methodological goals for the next four years are to develop a complete framework for estimation and inference of optimal adaptive treatment strategies and to develop tools specific to estimate such strategies using data from electronic health records. Questions of how best to personalize medicine are asked by patients experiencing a wide range of medical conditions; I will focus on determining optimal treatment sequences for two very different patient groups facing similar questions about treatment sequencing: those diagnosed with depression and those undergoing transplantation for leukemia.