The Philadelphia Department of Public Health (PDPH), in partnership with the Drexel Urban Health Collaborative, combined the 500 Cities data with other local health data to produce neighborhood-level health atlases and rankings across Philadelphia. These health atlases and rankings will be shared with local stakeholders and used to inform targeted public health efforts across the city.
HOW THEY USED 500 CITIES DATA
To begin their analysis of the 500 Cities data, the PDPH partnered with researchers from Drexel’s Dornsife School of Public Health to confirm data sources, develop neighborhood boundaries with geospatial analysis, and reaggregate the 500 Cities data based on new neighborhood boundaries.
The PDPH team first developed neighborhood boundaries with geospatial analysis, using the Southeastern Pennsylvania Household Health Survey and 2010 Census tract boundaries to inform geographic boundaries of 46 Philadelphia neighborhoods.
To document their work, the team created a detailed methodology—a technical guide that details their analytical approach and includes step-by-step guidance—that could help groups interested in creating similar neighborhood-level health rankings in their community and replicating their work. In this resource, the PDPH team showcases their approach to the following:
- defining neighborhood boundaries
- selecting neighborhood health measures and aggregating data for analysis
- standardizing and weighing health indicators
- creating composite scores
- ranking neighborhoods by health outcomes
The PDPH conducted their analysis on cleaned and verified source data, creating a collection of tables and maps that ranked each neighborhood across 500 Cities health outcomes. The team then compiled profiles of each neighborhood’s health indicators and final rankings in their Neighborhood Health Atlases and Rankings data interactive and final report.
WHAT THEY LEARNED
The Philadelphia Department of Public Health is excited about this project’s progress because it has been one of the first coordinated attempts to use intentionally crafted neighborhood profiles as a foundation of data analysis and public planning, as opposed to relying on predefined geographic boundaries coded within datasets. The PDPH looks forward to sharing these insights with the city of Philadelphia and starting a larger conversation about the importance of looking at data with special attention to geographic aggregation.
When analyzing the 500 Cities data, the PDPH found that many of the health outcomes are correlated. The team found that checking for variations across indicators is important when conducting multivariate analysis.
As the team moved through their analysis, they also learned that 500 Cities data includes estimates for all census tracts, including those with small populations. In creating their Neighborhood Health Atlases and Rankings, the Philadelphia Department of Public Health found that it may be more appropriate to exclude census tracts with smaller populations in some instances.
WHAT THEY RECOMMEND FOR SIMILAR PROJECTS
The Philadelphia Department of Public Health has identified the following recommendations for organizations interested in pursuing similar projects when using the 500 Cities data in their communities:
- Take time to develop geographical boundaries for analysis. Developing new geographical boundaries that also align with census tract–level estimates requires time and experts with an in-depth understanding of the geography.
- Acquiring data at the census-tract level from other local data sources can be a major barrier to replicating the ranking approach. Many local data sources typically produce estimates by other geographic boundaries (like zip codes) that don't directly correspond to census tracts. Ensure that your research team accounts for this analytical step when planning your project approach and crafting a project timeline.
To learn more about the Philadelphia Department of Public Health’s work, follow @PHLPublicHealth on Twitter. If you are interested in learning more about this team’s project approach or have specific questions regarding replicating their work in your community, please feel free to contact the Principal Investigator of this project, Raynard Washington.