About the speaker: Kevin B. Smith is a PhD candidate in the department of Industrial and Operations Engineering at the University of Michigan. His research is focused on applying data analytics tools to improve outcomes of problems in healthcare and public health policy. The mission of his research is to inform policymakers and the public of the critical factors that impact decisions with applications in organ allocation, pandemic risk planning, clinical kidney stone care, and AI/ML monitoring thus far. Additionally, he is motivated to develop models which improve decision making outcomes in these settings. His work aims to address systemic shortcomings of current healthcare and public health policies and create models that improve outcomes using interpretable and accessible results.

Abstract: At the intersection of the acceleration of the availability of data and efficient data analytic techniques is a unique moment to reconsider the way the US policy dictates community-level care delivery. Despite increased spending on novel treatments among other key innovations, shortcomings of the US’ policy delivery system have included poor outcomes of the country’s most vulnerable and the continuation of systemic barriers to better health. In this talk, I will demonstrate the use of data analytics to support these challenging policy contexts. In particular, I will present the results of a study wherein we developed a decision support tool designed to enable public health decision makers to rank US counties most at-need of scarce medical resources. I will also describe some opportunities and challenges we discovered when using publicly-available data and data analytics to guide decision making in this public health policy setting. I will conclude with a discussion of some other work supporting policy making decisions by using analytics on observational healthcare data and monitoring AI/ML models that are used in medical decision making settings.