Best Practices for the Design of COVID-19 Dashboards

Abstract

Since 2020, health informaticians have developed and enhanced public-facing COVID-19 dashboards worldwide. The improvement of dashboards implemented by health informaticians will ultimately benefit the public in making better healthcare decisions and improve population-level healthcare outcomes.

The authors evaluated 100 US city, county, and state government COVID-19 health dashboards and identified the top 10 best practices to be considered when creating a public health dashboard. These features include 1) easy navigation, 2) high usability, 3) use of adjustable thresholds, 4) use of diverse chart selection, 5) compliance with the Americans with Disabilities Act, 6) use of charts with tabulated data, 7) incorporated user feedback, 8) simplicity of design, 9) adding clear descriptions for charts, and 10) comparison data with other entities. To support their findings, the authors also conducted a survey of 118 randomly selected individuals in six states and the District of Columbia that supports these top 10 best practices for the design of health dashboards.

Keywords: health dashboard, health informatics, health information management, COVID-19, public health, data visualization

Introduction

A health dashboard is a visual display of health information used to highlight data for individuals and organizations for decision-making. Numerous types of health dashboards are accessible to the public for various diseases worldwide. These health dashboards provide individuals with essential information that can help increase safety, policies, and behavior. Several of the well-known dashboards include the Johns Hopkins COVID-19 dashboard, the Centers for Disease Control and Prevention (CDC) COVID-19 dashboard, the state of Maryland COVID-19 dashboard, and the Madison & Dane County COVID-19 dashboard.

A few of the authors presented their findings to the COVID-19 dashboard team in Montgomery County, Maryland. Through this presentation, the authors learned about several key elements essential to designing a COVID-19 dashboard at the county level. The authors continued their survey of COVID-19 dashboards by reviewing over 100 US city, county, and state government COVID-19 dashboards. Thereafter, the authors developed and performed a survey to help identify and confirm the top 10 best practices of COVID-19 dashboard design.

Background

The number of COVID-19 cases has rapidly increased over time. As of November 2022, there were over 98 million COVID-19 cases and over 1 million deaths in the United States alone. COVID-19 and long COVID has affected many individuals in countless ways. Although COVID-19 vaccines have recently been introduced, COVID-19 dashboards are still vital for everyone. The rampant infectious disease continues to spread despite the vaccine, and it is essential for everyone to be able to make healthcare decisions for themselves. The literature review on this topic found very few academic articles on the best practices for the design of COVID-19 dashboards and, even more importantly, health dashboards. 

Methods

The authors systematically examined over two years of data from 100 available US city, county, and state health dashboards related to COVID-19. The authors also provided recommendations to the COVID-19 dashboard team in Montgomery County, Maryland. Through this exercise and review of 100 COVID-19 dashboards, the authors identified 10 key design elements for public-facing health dashboards. Thereafter, the authors surveyed 118 individuals on COVID-19 dashboards to identify the top 10 best practices of any health dashboard. Demographic data were not collected for the first group of 58 individuals, but demographic data was collected for the second group of 60 individuals (Table 2). The 10 characteristics of creating and understanding COVID-19 dashboards design were incorporated and confirmed in the survey, and the results were collected (Table 1).

The authors conducted a survey of 118 (n=118) individuals above the age of 18. The survey consisted of 10 questions. Responses of “yes” counted as one point, while responses of “no” counted as zero points. The authors calculated a total of 1,181 responses, including 995 responses of “yes” and 186 responses of “no.” Incorporated user feedback had the lowest percent agreement of “yes” responses of 74 percent, whereas ADA compliance had the highest percent agreement of “yes” responses of 92 percent (Table 1).

Results

Based on the two surveys, the authors identified the top 10 design attributes of health dashboards (Figure 1) as noted below:

1. Easy Navigation

An essential feature of a dashboard is to allow the user to easily navigate through the various pages and elements of a health dashboard. Upon the authors’ review of 100 COVID-19 dashboards, it was noted that some were difficult to navigate from one page to another due to various navigation issues. Thus, several changes were implemented in terms of the location and size of the navigation features that are more in line with the current practices of other dashboards. In addition, a health dashboard may allow the user to hover over a data element to review additional information regarding that element, known as a “focus mode.”

2. High Usability

Another important element of a health dashboard is the ease of use. Key factors in usability are fast loading times, simple layouts, and readability when using the dashboard. Faster loading times, for example, allow for more user engagement, allowing the opportunity for people to look further into the data given in the dashboard. Furthermore, it is critical that the data is easily accessible, and the text is large enough for viewers to visualize the data clearly.

3. Use of Adjustable Thresholds

Interactive adjustable thresholds are an important attribute in dashboards because they can be beneficial to users to better understand the data. With adjustable thresholds, users can interact with the dashboard and change parameters for example from 14 days to 60 days for COVID-positive cases. By changing a parameter in a dashboard, the user can view the data in different ways that ultimately help them make better healthcare decisions. Additional thresholds for example based on percentages as opposed to numbers allow for more versatility for the user. This is more effective than a static dashboard, as the percentage of COVID cases decreasing during a certain period of time would inform the user of important, relevant data. One could even for example change from a monthly to a daily statistic to better understand a data trend (Figure 2).  

4. Use of Diverse Chart Selection

The use of multiple chart types is advantageous for a health dashboard. In order to increase the heterogeneity interest, diversity, and range in a health dashboard, it is important to provide different types of charts, graphs, and other data visualization options. Given the range of users that may visualize data, the addition of elements other than line charts that appeal to various groups, such as a heat map, will provide a variety to a dashboard. Studies show that users prefer to visualize different types of charts, such as bar charts or pie charts. Having graphs that interact with the viewers is helpful to have included so that they are capable of viewing the data from multiple different angles (Figure 3).             

5. Compliance with the Americans with Disabilities Act (ADA)

Given that data visualization often requires the use of color, it is important to ensure charts and graphs using color are compatible with the Americans with Disabilities Act (ADA). One in 12 men and one in 200 women are colorblind, and it is essential that everyone has an equal opportunity to fully understand the data being displayed. The most common form of color blindness includes the colors of red-green color blindness, and the avoidance of red and green should be considered. The use of standard colors that are ADA compliant is highly recommended.

6. Tabulation of Data Into Charts

Although it is essential to frequently demonstrate data and numbers in charts and graphs, it is equally important to avoid over-compression and summary of such data by providing the actual tabulated data. Furthermore, tabulating the data presented by charts allows easier access for individuals or companies seeking to use the statistics. Charts allow one to interpret the data in a different way, allowing one to visualize the data while tabulated data provides the actual numbers that may inform the user in a different way (Figure 4). 

7. Incorporated User Feedback

While a designer of any system will go out of their way to anticipate issues for the end user, it is essential to have an easy and simple way for users to provide feedback for a health dashboard. This feedback then must be incorporated into the health dashboard, as shown in Figure 5. Giving the option for users to leave remarks on a dashboard will immensely improve the effectiveness of a dashboard for the needs of individuals. 

8. Simplicity of Design

Users for any COVID-19 dashboard will have a range of education and the ability to interact with an online dashboard. Thus, based on the authors’ review of 100 dashboards and a survey of 118 adults, the complexity of displaying data should be minimized. This can be achieved by presenting data more simply and compactly while also providing layers where the most important information is displayed on the initial pages of the dashboard. Additional pages may allow the user to achieve greater levels of detail in reviewing dashboard data (Figure 6).

9. Adding Clear Descriptions

Based on the authors’ survey, they found that clear descriptions of the charts are necessary for the user to grasp a full understanding of the data being presented. Clearly describing what data is being shown in a chart or graph is imperative. Especially for certain people who don’t understand how charts and dashboard works, clear descriptions are vital. Clear labels for the x-axis and y-axis, for example, help the user better understand the data. Beyond charts, tables, figures, and other data, visualizations should be apparent to all users (Figure 7).

10. Comparison Data with Other Entities

When providing any type of data, it is crucial to allow it to be placed within a greater context. Thus, dashboards should have a feature to compare their data to similar entities. For example, Montgomery County data or any county’s data should be reviewed in the context of other counties so that a user can determine whether one county has fewer or greater cases than another. This gives users the ability to make healthcare decisions based on their area’s current situation as well as its surrounding situation (Figure 8).  

Discussion

The COVID-19 pandemic is the most impactful, significant, and researched event of our generation. COVID-19 is a disease process that potentially affects our entire population of over 7.8 billion worldwide, over 330 million in the United States, or simply 100,000 in the average county within the United States. It is essential that the best practices can be followed to enhance and optimize data visualization, as various cities, counties, and states develop dashboards for reporting COVID-19 data. The authors determined that 10 specific elements should be considered during the design of a COVID-19 dashboard. Further study is required to better validate the impact of each of these individual elements and to conduct a larger nationwide survey with demographic data. This study’s limitations include a limited geographic area in a few states.  

Conclusion

The COVID-19 pandemic is ongoing and continuously increasing in a number of cases and deaths. With the public seeking data regarding COVID-19 cases, deaths, vaccine usage, etc., accurate and effective dashboards are necessary for the public health information process. To ensure the dashboards are effective in the best ways possible, the authors have come to a conclusion with the 10 best practices to consider while creating a dashboard. These best practices will provide counties, states, and other designers with a set of guidelines for effective communication of COVID-19 statistics to the public. As these practices are in use, they will provide better health information while also improving public health.

Conflicts of Interest and Support

The authors have no conflicts of interest or financial support to report.

Notes

1. Ahn, June, Fabio Campos, Maria Hays, and Daniela DiGiacomo. “Designing in Context: Reaching beyond Usability in Learning Analytics Dashboard Design.” Journal of Learning Analytics 6, no. 2 (2019): 70–85. https://eric.ed.gov/?id=EJ1224124.
2. Barbazza, E, and D Ivankovic. “What Makes COVID-19 Dashboards Actionable? Lessons Learned from International and Country-Specific Studies of COVID-19 Dashboards and with Dashboard Developers in the WHO European Region.” European Journal of Public Health 31, no. Supplement_3 (October 1, 2021). https://doi.org/10.1093/eurpub/ckab164.488.
3. Barone, Stefano, Alexander Chakhunashvili, and Albert Comelli. “Building a Statistical Surveillance Dashboard for COVID-19 Infection Worldwide.” Quality Engineering 32, no. 4 (June 5, 2020): 754–63. https://doi.org/10.1080/08982112.2020.1770791.
4. CDC. “COVID Data Tracker.” Centers for Disease Control and Prevention, (March 28, 2020). https://covid.cdc.gov/covid-data-tracker/#datatracker-home.
5. Dong, Ensheng, Hongru Du, and Lauren Gardner. “An Interactive Web-Based Dashboard to Track COVID-19 in Real Time.” The Lancet Infectious Diseases 20, no. 5 (February 2020). https://doi.org/10.1016/s1473-3099(20)30120-1.
6. Few, Stephen. Information Dashboard Design: The Effective Visual Communication of Data. Sebastopol, Ca: O’Reilly, 2006.
7. Johns Hopkins Coronavirus Resource Center. “Johns Hopkins Coronavirus Resource Center,” 2020. https://coronavirus.jhu.edu/.
8. Lechner, Bettina, and Ann Fruhling. “Towards Public Health Dashboard Design Guidelines.” LNCS 8527 (2014): 49–59. https://link.springer.com/content/pdf/10.1007%2F978-3-319-07293-7_5.pdf.
9. Li, Veronica Qin Ting, and Masaru Yarime. “Increasing Resilience via the Use of Personal Data: Lessons from COVID-19 Dashboards on Data Governance for the Public Good.” Data & Policy 3 (2021). https://doi.org/10.1017/dap.2021.27.
10. Lyshol, Heidi. “7.C. Workshop: Dashboards for COVID-19: Lessons Learned.” European Journal of Public Health 31, no. Supplement_3 (October 1, 2021). https://doi.org/10.1093/eurpub/ckab164.483.
11. Martins, Nuno, Susana Martins, and Daniel Brandão. “Design Principles in the Development of Dashboards for Business Management.” Perspectives on Design II, (October 2, 2021). 353–65. https://doi.org/10.1007/978-3-030-79879-6_26.
12. Montgomery County Editors. “Data - COVID-19 Information Portal - Montgomery County, Maryland.” www.montgomerycountymd.gov, n.d. https://www.montgomerycountymd.gov/covid19/data/.
13. Mucchetti, Mark. “Dashboards and Visualization.” BigQuery for Data Warehousing, 2020, 379–400. https://doi.org/10.1007/978-1-4842-6186-6_17.
14. Nielsen, Jakob. Usability Engineering. Google Books. Morgan Kaufmann, 1993. https://books.google.com/books?hl=en&lr=&id=95As2OF67f0C&oi=fnd&pg=PR9&dq=Nielsen.
15. Sarikaya, Alper, Michael Correll, Lyn Bartram, Melanie Tory, and Danyel Fisher. “What Do We Talk about When We Talk about Dashboards?” IEEE Transactions on Visualization and Computer Graphics 25, no. 1 (January 2019): 682–92. https://doi.org/10.1109/tvcg.2018.2864903.
16. Taher, Faisal, John Hardy, Abhijit Karnik, Christian Weichel, Yvonne Jansen, Kasper Hornbæk, and Jason Alexander. “Exploring Interactions with Physically Dynamic Bar Charts.” Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems, (April 18, 2015). https://doi.org/10.1145/2702123.2702604.
17. Thorlund, Kristian. “A Real-Time Dashboard of Clinical Trials for COVID-19.” The Lancet Digital Health 0, no. 0 (April 24, 2020). https://doi.org/10.1016/S2589-7500(20)30086-8.
18. Ullrich, Alexander. “Dashboards as Strategy to Integrate Multiple Data Streams for Real Time Surveillance.” Online Journal of Public Health Informatics 11, no. 1 (May 30, 2019). https://doi.org/10.5210/ojphi.v11i1.9701.
19. Vahedi, Akram. “Applications, Features and Key Indicators for the Development of Covid-19 Dashboards: A Systematic Review Study.” Informatics in Medicine Unlocked 30 (2022): 100910. https://doi.org/10.1016/j.imu.2022.100910.
20. Wissel, Benjamin D. “An Interactive Online Dashboard for Tracking COVID-19 in U.S. Counties, Cities, and States in Real Time.” Journal of the American Medical Informatics Association 27, no. 7 (June 17, 2020): 1121–25. https://doi.org/10.1093/jamia/ocaa071.

Author Biographies

Dillon Malkani (dillon_malkani@bullis.org) is a student at Bullis School.

Melina Malkani is a student at Bullis School.

Neel Singh is a student at Landon School.

Eesha Madan is a student at BASIS Independent McLean.