Design as an Ingredient: Applying Human–Centered Design to Improve the Utility of Infectious Disease Scenario Projections for State, Tribal, Local, and Territorial Agencies (Preprint)

Résumé fourni par la source

BACKGROUND Infectious disease modeling tools are essential for public health decision-making, but their utility is limited when visualizations do not reflect end-user needs. The U.S.-based Scenario Modeling Hub (SMH) provides scenario projections for public health practitioners, yet previous designs were not informed by human-centered design principles. Applying these methods may improve clarity, interpretability, and real-world usefulness of model outputs for decision-makers. OBJECTIVE The goal of this study was to increase the relevance and utility of infectious disease models and scenario projections for public health officials nationwide by embedding human-centered design methods as an “ingredient” in the re-design of a modeling dashboard with visualizations. METHODS From January through July 2025, the Design and Innovation for the Public Good (DIPG) unit at The University of North Caroline at Chapel Hill, in partnership with the Atlantic Coast Center for Infectious Disease Dynamics and Analytics (ACCIDDA) and the Council of State and Territorial Epidemiologists (CSTE), facilitated three rounds of virtual focus groups with public health officials nationwide. Round 1 explored use cases for scenario projections; Round 2 gathered feedback on existing SMH visualizations and sketches; and Round 3 tested low-fidelity prototypes of the redesigned website. A structured assumption-testing framework guided prototype development, and data were synthesized using human–centered design (HCD) analysis tools. RESULTS Round 1 participants (n=19) identified seven priority information needs and emphasized needs for transparency around data sources and methods and actionable, locally tailored projections. In Round 2, participants (n=16) preferred simple, context-rich visuals such as epidemic curves, case counts, the ability to compare potential interventions, and curated insights. In Round 3, participants (n=9) supported comparing multiple scenarios, embedding takeaways within dashboards, and replacing mandatory onboarding with optional learning resources. CONCLUSIONS Embedding HCD as an “ingredient” can improve usability and acceptability of analytic tools for epidemiologists and public health officials nationwide so that these tools will be used more and be more helpful.