A Prototype Web Platform to Facilitate Public Engagement with Medical Evidence about Rheumatoid Arthtritis Medications

Abstract

Contemporary technologies and user interface design enable people to routinely interact with data in their everyday lives. While consumer applications for shopping and travel often feature data-driven user interfaces, health resources rarely do. These resources rely on manual translation of medical evidence into prose instead of providing users the capacity to interact with underlying data. The abstraction away from details about treatment options, including data about efficacy, harms, and patient-reported outcomes, stands in the way of people who may wish to become fully informed when taking on important medical decisions. In spite of barriers that restrict access to and potential to apply medical evidence, this project explored whether contemporary open-source Web technologies could be adapted to create datadriven resources for the exploration of such evidence. A prototype platform and example applications were developed using JavaScri+I3pt and React.js, with Google Spreadsheets as a data store for medical evidence related about twelve disease-modifying antirheumatic drugs (DMARDs) commonly used to treat rheumatoid arthritis. Research findings were manually encoded from diverse sources, and a controlled vocabulary and data visualization components built to bridge the gap between outcomes and data publishing formats favored in research, and issues important to patients with rheumatoid arthritis. The volume and heterogeneity of source evidence revealed no straightforward parallel to consumer data-driven online applications, especially where evidence conflicts or is uncertain. Nevertheless, this thesis demonstrates that extant and ready-made technologies can be combined to create an extensible, data-driven platform and user interface elements to investigate and visualize certain kinds of evidence about chronic disease treatment options. Future research might investigate how such platforms might be incorporated into patient-facing decision aids, automated synthesis of research findings, and collaborative tools to encode evidence.