Data Visualization Provides Unique Capabilities to Explain and Understand Complex Health-Care Issues

An important part of data science is visualization, which helps determine and understand the underlying patterns, trends, and correlations in datasets—information that may not be apparent in a spreadsheet.

A well-known example of applying data visualization in the health-care profession occurred nearly 170 years ago, during a cholera epidemic. London physician John Snow conducted an epidemiological study of the area and residents near the public water pump on Broad Street. By mapping the locations of the residences of those who died from the outbreak as well as their proximity to the pump, Snow determined that the pump was the source of the cholera epidemic that had infected the area.

The ongoing COVID-19 pandemic has clearly demonstrated that data visualization is a strategic tool in analyzing, predicting, and mitigating the spread of the disease. Visualization is also critical in conveying key information in formats that are easy to understand by both decision-makers and the general public.

Charles R. Drew University of Medicine and Science (CDU) is a well-respected institution located in Willowbrook, a community in Los Angeles County. CDU oversees residency training programs; allied health programs; a medical education program with the University of California, Los Angeles; and various centers for health disparities research.

Within CDU's College of Science and Health is the Department of Urban Public Health, which focuses on urban health disparities. Among the degrees offered by the department is a master in public health (MPH). One of the classes offered within the MPH course of study is Health Communication and Data Visualization, developed and taught by Fred Dominguez, MD, MPH.

"I began teaching data science at CDU in 2013 in a class entitled Health Education and Communication, which focused on public health literacy," says Dominguez. "We used various forms of media such as infographics and PowerPoint presentations as tools to present data for health findings to the community. In 2015, I introduced GIS (Geographic Information System technology) into the syllabus as an additional way to communicate health findings. Students used ArcMap desktop application. Based on their course evaluations for that year and each year following, students have noted that learning GIS was one of their class's highlights. Since 2019, I have switched and used only ArcGIS Online in the class because of its flexibility and easy access from different platforms."

"Some years ago," Dominguez continues, "I obtained a data science certificate and have worked for more than seven years as a research analyst for the Los Angeles County Department of Public Health (DPH) as well as teaching at CDU. At the DPH, I noticed that data analytics and geospatial analysis were becoming a primary focus for the public health workforce. Seizing on this trend, I wanted to better prepare our students with the skills needed for this emerging discipline. I believe that students should not only be familiar with the analytical capabilities of GIS but also (know) its ability to visualize and communicate results obtained from data analysis. So I changed the course focus from a public health literacy perspective to one emphasizing health data literacy and renamed it Health Communication and Data Visualization. Applying data literacy techniques in public health is about communicating health data findings through visualizations and storytelling, which also parallels to Esri's ArcGIS Business Analyst and ArcGIS StoryMaps products."

In the class, students choose a heath topic in Los Angeles County from the California Health Interview Survey database. The data is then exported, cleaned, and formatted. From the data, students create various explanatory visualizations—such as graphs, maps, and infographics—for an oral presentation using storytelling techniques. Students also learn how to digitally tell data stories through Esri's StoryMaps.

"Public health is very metric driven," says Dominguez. "The trained public health professional uses data in fields like epidemiology to describe health-related phenomena in populations. Another discipline, biostatistics, uses data applied to statistical methods to describe the significance of health-related phenomena. However, it is equally important that the results of their findings are communicated in an understandable way. The first step is to decide whether a visualization is needed or not in conveying the message. If the decision is to create a data visualization, then the question becomes, What kind of visualization? Data visualizations can be used for explanation, such as stacked bar charts and trend lines, or for exploratory purposes like dynamic dashboards. GIS maps, of course, can be used in both of these cases."

"A lot of the work we do in public health requires community and agency buy-in," says Carolina Rodriguez Orozco, a quality improvement coordinator at Behavioral Health Services, Inc., and a CDU MPH graduate. "Essentially, we need to be able to convince others that something is either a problem that needs to be addressed or that a program/intervention we have implemented is working. To be able to do this, we must understand what the data is telling us and present it in a compelling way so that our findings are understood and acted upon. Data literacy is crucial for this since [the data] can be easily misrepresented or misinterpreted. Without a foundational understanding of how the data was used to create a graph, chart, or map, health-care professionals cannot make informed decisions.

"Because location is often a significant component of data, using GIS opens a multitude of opportunities to visualize distributions or events, both locally and globally," Orozco continues. "Geospatial data has the added benefit of being interactive. You can add information to points of interest to provide a deeper insight about a specified location."

Challenges for public health professionals today include discovering health-related patterns and then developing and applying interventions that address disease detection, mitigation, and prevention.

"Location is a key component in understanding the distribution of public health patterns," Dominguez says. "The benefit of using GIS is its ability to analyze the related data, visualize their patterns, and determine where to allocate the necessary resources.

"Another challenge in public health is to encourage a change in behavior that can improve the quality of health for individuals and communities. Effective public messaging is critical in initiating this change," Dominguez concludes. "Creating simple informational and data-driven visualizations, particularly GIS maps, enhances the understanding of information, and with this understanding, behavioral change can occur."