Healthcare deserts: How accessible is US healthcare?
summary
Uses Medicare procedure volumes and ZIP lat/long to map every US ZIP to its nearest hospital, computes travel thresholds in Python (brute force geodesic first: 1 hr 25 min) and then with scikit BallTree using haversine distances (runs in ~4.5 seconds), exports scenarios into Tableau. Finds ~10 million without access at 25 miles and ~3.6M at 50 miles (policy changes drop affected population from ~5.8M to ~3.6M). Medicare procedure volumes BallTree haversine distances
AI-written summary · about
chapters (15)
- 0:00Introduction to session
- 1:03About Edwards Life Sciences
- 2:54Problem with Medicare policy
- 7:10Data sources and initial approach
- 12:07Optimizing distance calculations
- 16:47Using BallTree for speed
- 19:56Visualization best practices
- 21:42Broader applications of method
- 22:46Q&A session begins
- 23:06Impact on Edwards and patients
- 25:53Connecting with presenter
- 26:34Accessing Medicare data
- 29:47Who requested analysis
- 30:30Sample Python code offer
- 31:36Conclusion and Q&A wrap
related talks
35mPainting a Picture of Public Data
03h 59mGeospatial Public Policy Analysis with GeoPandas
40mHow to measure a city
37mHow quick can we be? (data visualization techniques for engineers)
41mHealthPlot: Understanding American public health through R
10mCivic Hacking & Urban Informatics: Mapping Data with GeoPandas
11m"Maps in Four Dimensions
13mRural Health Analytics with Oracle Health