A full-stack healthcare planning intelligence system that transforms Kenyan health facility, population, ownership, service availability, and KDHS 2022 indicator data into county-level planning insights.
The project helps answer one core question:
To what extent does the county-level distribution of healthcare facilities in Kenya reflect both public health need and healthcare market dynamics?
Instead of presenting raw facility records alone, the system turns fragmented public datasets into planning signals that can support county comparison, access review, service-gap analysis, ownership assessment, and health-need prioritization.
- Frontend: https://kenya-health-dashboard.vercel.app/
- County Explorer: https://kenya-health-dashboard.vercel.app/county-explorer
- Backend API: https://kenya-health-dashboard-api.onrender.com/
- API documentation: https://kenya-health-dashboard-api.onrender.com/docs
- GitHub repository: https://github.com/arapkirui513-hub/kenya-health-dashboard
Primary audience:
Data / Analytics roles
Secondary audience:
Health-tech / Digital Health roles
Supporting audience:
Full-stack development roles
Core portfolio message:
I build health data products that turn messy public datasets into practical planning intelligence.
Healthcare facility data is useful, but facility counts alone do not answer planning questions.
A county with many facilities may still face access pressure if its population is large. A county with fewer facilities may depend heavily on one ownership category. A county with better facility density may still show high health need based on maternal care, family planning, teenage pregnancy, or immunization indicators.
This project connects those layers into a planning workflow.
It helps users review:
- Facility distribution by county
- Population-adjusted facility access
- Public, private, faith-based, NGO, community, and academic ownership patterns
- Selected service availability
- ART service access
- County planning priority
- KDHS-based health need
- Side-by-side county comparison
- Printable county planning reports
- CSV exports for further analysis
The system supports questions such as:
- Which counties have the most health facilities?
- Which counties have fewer facilities relative to population size?
- Which counties rely more heavily on public facilities?
- Which counties show stronger private healthcare market activity?
- Which counties have faith-based or NGO-supported service delivery?
- Which counties show weaker selected service coverage?
- Which counties should planners review first?
- Which counties show higher KDHS-based health need?
- How do two counties compare across access, ownership, service coverage, planning priority, and health need?
Provides a national view of facility distribution, ownership patterns, service availability, access density, planning priority, and health need.
Calculates facility access relative to county population using facilities per 100,000 people.
This helps avoid overvaluing raw facility counts.
Shows how different provider categories shape healthcare access at county level.
Ownership categories include:
- Public
- Private
- Faith-based
- NGO
- Community
- Academic
The dashboard was optimized for production performance, accessibility, SEO, and lightweight initial loading.
Latest mobile PageSpeed Insights score:
| Category | Score |
|---|---|
| Performance | 95 |
| Accessibility | 100 |
| Best Practices | 100 |
| SEO | 100 |
| Agentic Browsing | 2/3 |
| Metric | Before | After |
|---|---|---|
| Performance | 56 | 95 |
| Accessibility | 100 | 100 |
| Best Practices | 100 | 100 |
| SEO | 100 | 100 |
- Added route-level code splitting for Dashboard, County Explorer, and Map pages.
- Lazy-loaded heavy dashboard sections.
- Deferred Facility Finder API calls until users scroll near that section.
- Reduced the Dashboard JavaScript chunk from about 435 kB to about 48 kB.
- Prevented the large MapPage bundle from loading during homepage initial load.
The PageSpeed result is documented in:
docs/PERFORMANCE_OPTIMIZATION_NOTE.mddocs/screenshots/pagespeed-after-performance-optimization.png
Ranks counties using planning signals from:
- Access risk
- Service risk
- Ownership risk
- Population pressure
The index produces:
- Priority score
- Priority level
- National rank
- Component risk scores
- Reason flags
Uses KDHS 2022 county indicator data to estimate health need across:
- Teenage pregnancy
- Family planning need
- Maternal care gaps
- Child immunization gaps
The index produces:
- Health need score
- Health need level
- Component scores
- Input metrics
- Reason flags
Allows side-by-side comparison of two counties across:
- Facility access
- Population-adjusted access
- Ownership mix
- Service coverage
- Planning priority
- Health need
V4.0.2 added planning-grade reporting support:
- Printable county comparison report
- Planning Priority CSV export
- Priority-level filters
- Methodology modal
- Improved loading states
- Improved error states
- Improved empty states
The project combines:
- Kenya health facility records
- 2019 Kenya county population data
- KDHS 2022 county-level indicator data
Data work includes:
- County-name normalization
- Facility ownership grouping
- Service availability parsing
- Population-adjusted density calculations
- County-level indicator integration
- Derived planning index creation
- API-safe output formatting
The access layer calculates:
- Total facilities per 100,000 people
- Public facilities per 100,000 people
- ART facilities per 100,000 people
This helps compare counties more fairly than raw facility counts alone.
The Planning Priority Index combines:
Priority Score =
Access Risk * 0.40
+ Service Risk * 0.30
+ Ownership Risk * 0.20
+ Population Pressure * 0.10
Priority levels:
High: 70 and above
Medium: 40 to 69.99
Low: below 40
Full methodology:
docs/PLANNING_PRIORITY_INDEX_METHODOLOGY.md
The Health Need Index uses KDHS 2022 county indicators to estimate relative health need.
The current backend calculates component scores for:
- Teenage pregnancy risk
- Family planning need risk
- Maternal care gap risk
- Child immunization gap risk
Full methodology:
docs/HEALTH_NEED_INDEX_METHODOLOGY.md
Key endpoints include:
GET /
GET /health
GET /summary
GET /ownership
GET /facility-types
GET /counties
GET /services
GET /service-gap-score
GET /population
GET /access-density
GET /planning-priority-index
GET /kdhs-indicators
GET /county-health-needs
GET /health-need-index
GET /need-access-gap-index
GET /facilities
GET /facilities/export
API documentation:
https://kenya-health-dashboard-api.onrender.com/docs
- React
- Vite
- JavaScript
- Recharts
- CSS
- FastAPI
- Python
- pandas
- SlowAPI rate limiting
- CSV export support
- Frontend: Vercel
- Backend: Render
- API docs: FastAPI OpenAPI / Swagger UI
Answered:
Where are Kenya’s health facilities, and how are they distributed?
Included:
- Facility distribution dashboard
- County-level facility counts
- Ownership analysis
- Service availability
- ART service gap analysis
- Facility search and filters
- CSV export
- FastAPI backend
- React frontend
- Public deployment
Answered:
Are facilities distributed according to population and ownership patterns?
Included:
- County population dataset
- Population-adjusted facility access
- County Explorer
- Side-by-side county comparison
- County Insight Brief
- Ownership & Market Dynamics section
- Mobile polish
- Backend security improvements
Answered:
Which counties should planners pay attention to first?
Included:
- Planning priority endpoint
- Access risk scoring
- Service risk scoring
- Ownership risk scoring
- Population pressure scoring
- Priority levels
- Reason flags
- Planning Priority dashboard section
Answered:
Which counties show higher health need based on KDHS 2022 indicators?
Included:
- KDHS 2022 county indicator layer
- Health Need Index
- Health need levels
- Component scores
- Reason flags
- Health Need dashboard section
- County Explorer health-need comparison
Answered:
Can users export, explain, and share the planning insights?
Included:
- Planning-grade County Explorer print report
- Planning Priority CSV export
- Priority-level filters
- Methodology modal
- Improved loading states
- Improved error states
- Improved empty states
Status: Released (Frontend V5 • Backend API v5.0.2)
V5 extends the Kenya Health Facilities Dashboard with a Need vs Access Gap layer that combines healthcare need indicators with service access metrics to identify counties where healthcare demand is high relative to available services.
Key additions include:
- Need vs Access Gap Index
- County intervention prioritization
- County Explorer integration
- Backend API support (
/need-access-gap-index) - Frontend visualization and interpretation
- Production-ready portfolio release
See docs/V5_PRODUCTION_PORTFOLIO_RELEASE.md for implementation details and release notes.
- V1 completion report:
docs/V1_COMPLETION_REPORT.md - V2 completion report:
docs/V2_COMPLETION_REPORT.md - V3 completion report:
docs/V3_COMPLETION_REPORT.md - V4 completion report:
docs/V4_COMPLETION_REPORT.md - V4.0.2 completion report:
docs/V4_0_2_COMPLETION_REPORT.md - V5 production release:
docs/V5_PRODUCTION_PORTFOLIO_RELEASE.md - Security review:
docs/SECURITY_REVIEW.md - Planning Priority methodology:
docs/PLANNING_PRIORITY_INDEX_METHODOLOGY.md - Health Need methodology:
docs/HEALTH_NEED_INDEX_METHODOLOGY.md
This project demonstrates applied data and product skills across:
- Public dataset cleaning
- Healthcare analytics
- County-level planning analysis
- Feature engineering
- Index design
- API development
- Frontend data storytelling
- Documentation
- Deployment
- Planning workflow design
It shows how messy public health data can become a practical product layer for review, comparison, and planning.