End-to-end healthcare analytics portfolio demonstrating claims analytics, population health analysis, patient risk stratification, and executive dashboards using Python.
Population Health Workbench is a healthcare analytics portfolio project built using synthetic healthcare data generated by Synthea. It demonstrates how healthcare claims and clinical data can be transformed into actionable insights using Python through analytical workflows commonly used by healthcare payers, provider organizations, accountable care organizations (ACOs), and healthcare consulting teams.
The project integrates claims, patient demographics, encounters, and clinical conditions to develop interactive dashboards, analyze utilization patterns, stratify patient risk, and identify opportunities for population health interventions and healthcare payment strategy.
The solution emphasizes reusable Python modules, interactive Plotly visualizations, and business-oriented interpretation to support data-driven decision-making rather than exploratory analysis alone.
- Analyze healthcare claims and utilization patterns
- Evaluate population health characteristics
- Stratify patients by clinical and financial risk
- Identify opportunities for care management
- Demonstrate modular healthcare analytics using reusable Python code
- Present findings through executive dashboards and business interpretations
Explore each notebook for detailed analyses, interactive visualizations, and business interpretations.
| Notebook | Description | Screenshots |
|---|---|---|
| 📊 01. Data Exploration | Exploratory analysis of the Synthea dataset. | — |
| 📈 02. Executive Dashboard | Executive summary of key healthcare metrics. | Images |
| 💰 03. Claims Analytics Dashboard | Claims, financial, provider, and utilization analytics. | Images |
| 🏥 04. Population Health Analytics | Population health, risk stratification, and care management analyses. | Images |
The notebooks are powered by reusable analytics modules located in the src directory.
| Module | Purpose |
|---|---|
| 📄 data_loader.py | Loads and prepares healthcare datasets. |
| 📄 claims_analytics.py | Functions for claims, financial, provider, and utilization analytics. |
| 📄 population_health.py | Functions for prevalence analysis, risk stratification, and population health metrics. |
| 📄 terminology.py | Helper functions for clinical terminology and healthcare data processing. |
PopulationHealthWorkbench/
│
├── notebooks/
│ ├── 01_Data_Exploration.ipynb
│ ├── 02_Executive_Dashboard.ipynb
│ ├── 03_Claims_Analytics_Dashboard.ipynb
│ └── 04_Population_Health.ipynb
│
├── src/
│ ├── claims_analytics.py
│ ├── data_loader.py
│ ├── population_health.py
│ └── terminology.py
│
├── data/
│ └── raw/
│
├── images/
│
└── README.md
- Python
- Pandas
- NumPy
- Plotly
- Jupyter Notebook
- Population summary
- Financial overview
- Key healthcare metrics
- Interactive executive visualizations
- Claim volume analysis
- Diagnosis frequency
- CPT utilization
- Provider activity
- Healthcare expenditures
- Business insights
- Condition prevalence
- Patient utilization
- Population risk stratification
- Risk tier analysis
- Care management opportunities
- Executive recommendations
Healthcare Data
│
▼
Data Loading
│
▼
Reusable Python Analytics Modules
│
▼
Interactive Plotly Dashboards
│
▼
Business Interpretation
│
▼
Executive Recommendations
- Healthcare Analytics
- Population Health
- Claims Analytics
- Risk Stratification
- Data Visualization
- Business Intelligence
- Python Programming
- Data Cleaning
- Modular Software Design
- Executive Reporting
This project uses synthetic healthcare data generated by Synthea.
No real patient information or protected health information (PHI) is included.
Potential future additions include:
- Quality Measures Dashboard
- Value-Based Care Analytics
- Provider Performance Analytics
- Predictive Risk Modeling
- Streamlit Web Application
This repository is intended for educational and portfolio purposes only.
The analyses use synthetic healthcare data and a demonstration rule-based risk stratification approach. They should not be used for clinical decision-making.
Divya Verma, MS (Health Informatics), RHIA
Healthcare Analytics • Health Informatics • Medical Necessity • Python • SQL