The push for open data inside Philly’s Department of Public Health

Since that flashy Community Health Explorer tool dropped last month, some 2,000 users have tinkered around with the 77 datasets available on the platform, according to city officials.
Although there’s no number available, Dr. Cheryl Bettigole, Division Director for Chronic Disease Prevention at the Department of Public Health (PDPH), said there has been an increased interest in the data the PDPH had previously been sitting on — in the form of dreaded PDFs and slide decks.
See, since 2014, the department has been putting out information on public health indicators in the form of Community Health Assessment reports. Granted, they weren’t in the friendliest of formats (here, have a 132-page PDF).
“People would have to pass through hundreds of slides to get the information,” Bettigole admits.
So when the Office of Open Data and Digital Transformation first mentioned the concept of the Community Health Explorer, the PDPH was all over it.
“This is a tool we’ve been very excited about and working hard on for close to a year,” said Bettigole.
And then Amory Hillengas and Meagan Pharis came into the picture, who Lauren Ancona — the city data scientist who spearheaded the project— credited as instrumental parts of the tool’s development.
“The biggest process is completing the annual report,” Pharis, a data analyst for PDPH, said of working on the project. “But as for translating the report to this online interactive platform, the work was more about data integrity, about making sure it was presented accordingly.”

Although all an average user sees while tinkering around with the Community Health Explorer are smooth graphics and charts, you can imagine where all that flashy info came from: boring, wonky spreadsheets.
The team took on the task of cleaning up extensive Excel files with all the raw data before it could be turned into the final product.
https://twitter.com/PHLPublicHealth/status/765867268710555648
“We had a lot of notation and extra noise in our spreadsheets and it wasn’t as translatable to the external audience,” said Pharis.
The biggest lesson for Pharis was being able to tell the difference between how a health data scientist views data and how developers view it, and jumping across those differences in order to translate between shorthands.
“And also, learning how important data visualization is for getting your message out there, explains Pharis. “We get more feedback and more notice in this format.”
But in the end, all the usable, searchable data in the world is no good unless it gets used and generates action. That’s why, for Bettigole, the ideal interaction between the explorer and the average citizen goes like this:
“In my ideal world, people would look at the tool and have conversations with friends and family,” she said. “They would ask questions like, ‘If diabetes rates in my neighborhood are high, why is that?'”

---