Health Resilience Mapping: Finding Communities That Defy the Odds

Identifies US census tracts with unexpectedly good health outcomes despite limited food access by analyzing 68,000+ tracts linking CDC PLACES health data with USDA Food Access Atlas.

Table of Contents

• Key Findings
• Live Demo
• Documentation
• Quick Start
• Data Sources
• Project Structure
• Contributing
• Citation
• License

Key Findings

The Discovery

Our analysis of 68,170 census tracts across all 50 US states reveals 1,059 communities (1.6%) that demonstrate exceptional health resilience despite being classified as Low-Income Low-Access (LILA) areas. These "resilience hot spots" exhibit health outcomes 0.6-4.7 standard deviations better than predicted.

Top Resilient Communities

Rank	Location	County	Resilience Score
1	Tennessee 47149041500	Rutherford	4.75
2	South Carolina 45077011202	Pickens	4.41
3	South Carolina 45013001000	Beaufort	4.32
4	Michigan 26107981300	Mecosta	4.24
5	Kentucky 21227010400	Warren	4.22

Geographic Patterns

• Southeast clustering: Strong resilience patterns in rural South
• Midwest industrial cities: Pockets of unexpected health outcomes
• State leaders: Indiana, South Carolina, Tennessee show highest concentrations

Potential Protective Factors

Factor	Description
Social Capital	Strong community bonds and support networks
Faith-Based Infrastructure	High church density correlating with resilience
Alternative Food Systems	Gardens, farmers markets, informal economies
Healthcare Access	Presence of FQHCs and mobile clinics
Cultural Practices	Traditional foodways and community resilience strategies

Live Demo

View the Interactive Map - Explore all 68,170 census tracts with filtering, search, and detailed community profiles.

Documentation

Research

Document	Description
Research Findings	Complete statistical analysis and results
Research Paper	Academic manuscript in preparation
Policy Analysis	Implications for health equity initiatives
Methodology	Detailed methodology and validation

Development

Document	Description
Technical Architecture	System design and infrastructure
Development Setup	Local development guide
Roadmap	Planned features and improvements

Quick Start

Prerequisites

• Node.js 20+
• Python 3.8+ (for analysis scripts)
• Go 1.21+ (for API development)

Web Application

# Clone the repository
git clone https://github.com/cschuman/resilience-mapping.git
cd resilience-mapping

# Install dependencies and run dev server
cd app/web
npm install
npm run dev

Python Analysis

cd app/analytics

# Set up Python environment
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

# Run comprehensive analysis
python analyze_resilience.py

Data Sources

Source	Year	Records	Description
CDC PLACES	2023	2.55M	Census tract health outcomes
USDA FARA	2019	72,531	Food access indicators
Census TIGER/Line	2020	-	Tract boundary shapefiles

Important Caveats

• Temporal Gap: 4-year difference between FARA (2019) and PLACES (2023) data
• Geographic Boundaries: Mixed 2010/2020 census tract definitions
• Ecological Inference: Tract-level patterns don't imply individual behaviors
• Model Estimates: PLACES uses model-based estimates, not direct measurements

Project Structure

resilience-mapping/
├── app/
│   ├── web/                 # SvelteKit application
│   │   ├── src/routes/      # Pages and API endpoints
│   │   ├── src/lib/         # Components and utilities
│   │   └── fly.toml         # Fly.io deployment config
│   ├── analytics/           # Python analysis scripts
│   └── scripts/             # SQL schema and utilities
├── data/
│   ├── input/               # Source data (CSV, shapefiles)
│   └── output/              # Generated results and figures
├── docs/                    # Research and development documentation
│   ├── research/            # Research papers and findings
│   ├── architecture/        # Technical documentation
│   └── development/         # Development guides
└── .github/                 # CI/CD workflows and templates

Contributing

We welcome contributions! See CONTRIBUTING.md for guidelines.

Areas for Contribution

• Incorporating additional social determinants data
• Temporal analysis with multiple years
• Machine learning approaches for pattern detection
• Qualitative validation through community interviews
• Accessibility improvements

Citation

If you use this analysis in your research:

@software{resilience_mapping_2025,
  author       = {Schuman, Corey},
  title        = {Health Resilience Mapping: Finding Communities That Defy the Odds},
  year         = {2025},
  publisher    = {GitHub},
  url          = {https://github.com/cschuman/resilience-mapping},
  note         = {Analysis of 68,170 US census tracts identifying 1,059 communities
                  with exceptional health resilience despite limited food access}
}

License

MIT License - See LICENSE file for details.

Acknowledgments

This analysis builds on publicly available data from:

• Centers for Disease Control and Prevention (CDC)
• United States Department of Agriculture (USDA)
• U.S. Census Bureau

---

Report Bug · Request Feature · Discussions