London, circa 1854. A deadly cholera outbreak has caught the neighborhood of Soho in its teeth. The young, the old, the hale, the weak—this is an indiscriminate disease, and it’s moving fast. Hundreds die in little more than a week. The medical community races to provide solutions, but the best scientific methods of the day fail to have an impact.
There’s no miracle cure in this story. Instead there’s a young doctor named John Snow, and a radical idea. Thirty years before the world would read about a certain Mr. Holmes, Snow appoints himself a detective and heads into the heart of Soho to collect, integrate, and analyze data related to cholera deaths, pioneering the concept of epidemiology along the way. He plotted his findings on a map of Soho (see Figure 1), which persuaded officials to disable a public water pump that may have been a source of the outbreak. At about the same time, cholera deaths slowed and then stopped.
FIGURE 1: John Snow’s 1854 cholera map
Snow’s actions seeded a practice that would become critical in resolving future pandemics: data integration and visualization. Although he couldn’t have imagined the powerful software we’d be using nearly two hundred years later, he’d surely recognize its aim. If we can turn a seemingly chaotic event into data, and turn that data into dynamic visualizations, then we have the means to recognize the pattern of a virus or illness to help arrest its progress. This is what John Snow did in 1854, and it’s what we’re doing today:
- First, process large volumes, types, and sources of data
- Next, translate that data into understandable, informative visuals
- Finally, enrich and interpret that data with context-sensitive annotations
How Integrated Data Visualizations Inform Decisions and Impact Actions
Many decades later, the academic and statistician Edward Tufte would praise Snow as the agent of “a life-saving scientific discovery,” and would hold up the cholera map as an early example of how data integration and visualization can guide action. Tufte pointed to what he calls the map’s “supplemental details”— that is, how Snow uses “image, word and number… to present the evidence and make the argument.” Through these annotations, Snow had elevated the map’s meaning. It wasn’t simply a network of streets, sketched over with symbols; Snow had fashioned it into a powerful tool for identifying the source of a health threat, neutralizing that threat, and ultimately saving lives.
Data Visualization as a Tool to Fight Modern Pandemics
Just as Snow’s work helped to save lives in 1854, our modern data visualizations can help save lives today by making the “invisible” threat of SARS-CoV-2 more tangible and better understood.
To do that, we need more than inert data dashboards. The scale of this pandemic and the diversity of data surrounding it make this a uniquely complex situation which requires dynamic tools to unpack. You can see an example of such a dynamic tool in Figure 2, which demonstrates how PerkinElmer’s COVID-19 Analytics Resource Center, built by a team of life science and data science experts, takes viewers from a highly aggregated state-level view of the pandemic to county-specific information in just a click. Like opening the layers of a nested doll, the software allows us to penetrate layer after layer of quantitative detail, discovering new details that lie within the aggregated whole.
FIGURE 2: This COVID-19 Visual Analysis Hub allows users to drill into county-specific statistics, such as this report from Los Angeles.
Those details, hidden inside the statistics splashed across news headlines are key. Tufte explained it best when he wrote that “the credibility of [Snow’s] cholera map grows out of supplemental details in the text.” To give our narrative around COVID-19 equal credibility, we need to become devotees of those “supplemental details.” Without these annotations, a certain degree of meaning and insight would drain away, leaving only an alarming set of statistics with little to explain or contextualize what’s really going on. Figure 3, which depicts daily COVID-19 cases in South Korea, shows how these supplemental details contribute to a more complete picture of the pandemic, as just one example.
FIGURE 3: Users can toggle between different data visualizations for a more nuanced picture of the pandemic's impact.
The bottom line: data visualization is important, but as Snow intuited in 1854 and as we know from two centuries of practice, every visualization needs annotations and details to give it meaning. That’s what the best COVID-19 data platforms are designed to provide: an accurate and up-to-date picture of the quantitative numbers, along with the supplemental details that scientists need to help them take appropriate action to help fight this pandemic.
Pulling Meaning From the Madness: A Final Word
When John Snow got to work in 1854, all he had was a notebook and his own two feet. Today, we have intelligent analytics engines that can integrate data from disparate sources and present it in compelling, meaningful visuals, accurately and in an instant. And while Snow used his annotated map for very specific means — to prove his theory and potentially save more lives — today’s powerful data platforms have endless use cases. It’s that very flexibility that makes modern analytical tools so valuable. Whether investigating the effectiveness of a promising drug candidate, or monitoring clinical trial data for safety signals, the right data platform provides the insight that we need to accelerate our pandemic response.
In 1854, John Snow used data visualization to help people make informed and impactful decisions at the level of a single community up through countries. Today, we’re using powerful analytic and visualization software to apply Snow’s premise on a global scale, helping to integrate a huge variety of data quickly, accurately, and in truly meaningful ways. To learn more visit our COVID-19 Analytics Resource Center.