How Johns Hopkins’ coronavirus dashboard came to beMarch 17, 2021
In the early days, the Hopkins team manually gathered COVID-19 data that had been shared through “nontraditional” outlets, Gardner says.
“Data was coming from posts on things like Twitter and Facebook, as well as some one-off news articles,” she said. There weren’t any “standing authoritative websites” that were providing this kind of public health data.
So the team at Hopkins decided to make the data it was collecting public and to incorporate an interactive map where people could track the spread of cases around the world. In January 2020, soon after the first reported COVID-19 case in the US, the university shared the information it had gathered on Twitter. It was quickly picked up and amplified by news outlets.
On the first day of being publicly available, the dashboard hit a few thousand pageviews. By the end of the first week, it had reached around 10 million views. And in April, as the virus spread rapidly, the dashboard reached over 4 billion hits a day.
The dashboard has become an important source for elected officials, public health leaders and news outlets,, to pull COVID-19 data including the total number of cases around the world, cases by country or region and global deaths. It’s been used to inform policy decisions, according to Hopkins. Since March, the university’s COVID sites have had more than 476 million visitors, and have garnered 1 billion page views.
Not surprisingly, there were some challenges when it came to data collection for the dashboard. Gardner says they had to deal with the rapid growth of the outbreak and the increasing demand being placed on the infrastructure through which they were providing data. They had to identify leading authoritative sources to pull data from, and figure out how to automate data collection from them. After reliable sources were identified (including the World Health Organization and the US Centers for Disease Control and Prevention), the data would still come in in various formats, meaning it had to be cleaned and processed. It became too much to manage with the initial manual data collection and entry processes in place, so Hopkins implemented an automated process.
Aaron Katz, supervisor of the large-scale analytic systems group at Johns Hopkins’ Applied Physics Laboratory, was part of the team that automated the dashboard’s data collection and management. He says the dashboard now runs on a mostly autonomous data pipeline, with global data being pulled throughout the day. The team collects roughly 40,000 overlapping data points an hour, which produce the nearly 4,000 that exist in the public data set.
Katz and Tamara Goyea, senior data scientist at the Applied Physics Lab, say they hope these kinds of technical approaches will be used for more public health challenges in the future.
Beth Blauer, executive director of the Centers for Civic Impact at Johns Hopkins, added that this kind of approach and technology could also be used to tackle concerns like climate change, violence and food insecurity.
“I hope that we take advantage of this infrastructure in a way where we can start to jointly tackle some of these real key critical issues that have been playing in our local communities for generations,” Blauer said.
The information contained in this article is for educational and informational purposes only and is not intended as health or medical advice. Always consult a physician or other qualified health provider regarding any questions you may have about a medical condition or health objectives.