(SLA-066) Landscape of Public Health Data Dashboards: Comparison of Key Features for Epidemiology Research

Background: Digital data dashboards (DDDs) are web-based tools that collate, analyze, and visualize timely data from disparate sources. They gained popularity during the pandemic as COVID-19 underscored the critical need for online real-time, map-based data to monitor disease trends and inform mitigation strategies.

Objectives: Our aim was to describe and compare key attributes of public population-level healthcare (HC) DDDs with a major epidemiology research database (DB).

Methods: Keyword searches were conducted in Google/Scholar, prioritizing dashboards that were ranked highly in search results, are publicly accessible, span diverse health domains, and cover different regions. Dashboards were characterized by name, therapeutic area, region, organizer, aim, key indicators, data sources, methods, time focus, visual & interactive properties, download options, and limitations. Two separate public dashboards were then compared with an epidemiology research DB, as profiled in B.R.I.D.G.E. TO DATA® (www.bridgetodata.org).

Results: In total 32 DDDs were identified, covering (not mutually exclusive): COVID-19 (12), influenza (1), HIV/AIDS (2), healthcare (5), drug prescribing (2), drug abuse (3), cancer (1), maternal-infant health (2), and global indicators (8). Most dashboards aggregate data from multiple sources (25) and stratify by geography (33). Nine dashboards (COVID-19 & influenza) provide real-time updates; vaccine-related dashboards further stratify doses administered, product, at-risk groups, income group and estimated uptake rates. While 22 dashboards define specific health outcomes, their validation processes vary. Attributes of Our World in Data – COVID-19 and Illinois Opioid Data Dashboard were compared to B.R.I.D.G.E.’s profile of the Clinical Practice Research Database (CPRD). Shared features include large DB sizes, inpatient & outpatient data, diagnoses, and death counts. DDDs differ from CPRD in accessibility, analytical scope (trends vs. longitudinal patient-level analyses), and granularity. Both DDDs lack sociodemographic, behavioral, and cost data, while disease mitigation data (e.g., policy measures) are unique to these DDDs.

Conclusions: Data dashboards supported surveillance and response efforts by local and global health departments even prior to the COVID-19 pandemic. Common attributes of these DDDs include geographical stratification, disease mitigation tracking, and interactive visualization tools. To increase awareness and enhance research usability, B.R.I.D.G.E. can serve as a standardized framework offering structured descriptions of HC DDDs.