Features Constituting Actionable COVID-19 Dashboards: Descriptive Assessment and Expert Appraisal of 158 Public Web-Based COVID-19 Dashboards

Background Since the outbreak of COVID-19, development dashboards as dynamic, visual tools for communicating COVID-19 data has surged worldwide. Dashboards can inform decision-making and support behavior change. To do so, they must be actionable. The features that constitute an actionable dashboard in context pandemic have not been rigorously assessed. Objective aim this study is to explore characteristics public web-based by assessing their purpose users (“why”), content (“what”), analyses displays (“how” communicate data), ultimately appraise common highly dashboards. Methods We conducted a descriptive assessment scoring using nominal group technique with international panel experts (n=17) on global sample July 2020. sequence steps included multimethod sampling dashboards; piloting tool; extraction initial round actionability scoring; workshop based preliminary analysis results; reconsideration scores followed joint determination used statistics thematic findings research question. Results A total 158 from 53 countries were predominately developed government authorities (100/158, 63.0%) national (93/158, 58.9%) scope. found only 20 (12.7%) stated both primary intended audience. Nearly all reported epidemiological indicators (155/158, 98.1%), health system management (85/158, 53.8%), whereas social economic impact behavioral insights least (7/158, 4.4% 2/158, 1.3%, respectively). Approximately quarter (39/158, 24.7%) did report sources. time trends disaggregated two geographic levels age sex. average 2.2 types (SD 0.86); these mostly graphs maps, tables. interpretation, color-coding was 89.4%), although one-fifth (31/158, 19.6%) text explaining quality meaning data. In total, 20/158 appraised actionable, seven identified between them. Actionable (1) know audience information needs; (2) manage type, volume, flow displayed information; (3) sources methods clearly; (4) link policy decisions; (5) provide are “close home”; (6) break down population into relevant subgroups; (7) use storytelling cues. Conclusions diverse why, what, how which data-driven decision-making. leverage full potential, developers should consider adopting identified.