Ensuring patient privacy and accuracy of analytical methods to support evidence-based healthcare

Abstract

Over the past two decades, healthcare providers substantially increased their use of electronic health record (EHR) systems. EHRs are primed to become the core of the data driven healthcare system, with the potential to serve as a platform for population health analytics, predictive model development and implementation, and coordination with patients to manage their health information. However, research with EHRs introduces the risk of exposing patient records and business practices to nefarious actors. Creating infrastructure to deliver predictive methods to clinical records while protecting patient privacy is key to building a reliable healthcare analytics platform. In addition, the quality of data from these systems is not fully validated for all use cases, such as assessing population health. Validating the utility of EHRs for use as a population health platform is necessary to fully realize the vision of the data driven health system. Patient involvement in their health is essential to maximize positive patient outcomes. While many vectors exist for patients to access their health information, they are still limited in their ability to contribute to their health data. More solutions are needed to further promote patient involvement with their healthcare information. In this dissertation, I focus on three areas with four aims for building a safe, private, and accessable data analytics platform on the EHR. The aims are to: (1) Evaluate the University of Washington EHR as a generalizable public health repository; (2) Pilot a "Model to data" framework as a method to deliver predictive analytic methods to clinical records; (3) Scale the "Model to data" pipeline to host a community challenge, securely delivering outside models to EHRs; and (4) Develop a patient portal to enable patientinteraction with their health data and the return of clinically actionable research results.