Intelligent formulary change detection and analysis system for healthcare organizations.

🔗 Try it now →

Upload your own formulary files and see instant analysis. No installation required!

(Deploy your app first, then replace the URL above with your actual Streamlit URL)

Visit the live demo above - upload files and analyze instantly!

# Clone repository
git clone https://github.com/your-username/ai-formulary-intelligence.git
cd ai-formulary-intelligence

# Install dependencies
pip install -r requirements.txt

# Launch Streamlit app
streamlit run app.py

Open http://localhost:8501 in your browser.

# Analyze single carrier
python src/formulary_delta_processor.py february.csv \
    --carrier "CarrierName" \
    --keys RXCUI

# Batch process multiple carriers
python src/batch_processor.py data/

# Generate SQL for database loading
python src/generate_sql.py \
    --carrier UHC \
    --timestamp 20240215_103045 \
    --keys RXCUI

• ✅ Upload and compare formulary files
• ✅ Automatic change detection (added/deleted/modified)
• ✅ Support for multiple file formats (pipe, CSV, TSV)
• ✅ Real-time analysis (processes 500K+ records in seconds)
• ✅ Export results as CSV
• ✅ Download summary reports

• ✅ Batch processing for multiple carriers
• ✅ Historical tracking and archiving
• ✅ SQL script generation for database loads
• ✅ Detailed change categorization
• ✅ AI-powered insights (optional with Claude API)

• Formulary additions and deletions
• Tier changes (1→2, 2→3, etc.)
• Prior Authorization (PA) updates
• Step Therapy (ST) requirements
• Quantity Limit (QL) modifications
• Copay/coinsurance changes
• Specialty designation updates

┌─────────────────────────────────────────┐
│     Interactive Web Interface          │
│         (Streamlit App)                 │
└─────────────────────────────────────────┘
                  ↓
┌─────────────────────────────────────────┐
│      Analysis Engine (Python)           │
│  • Multi-format parser                  │
│  • Delta detection                      │
│  • Change categorization                │
└─────────────────────────────────────────┘
                  ↓
┌─────────────────────────────────────────┐
│    Output & Integration Layer           │
│  • CSV exports                          │
│  • SQL generation                       │
│  • Summary reports                      │
└─────────────────────────────────────────┘

ai-formulary-intelligence/
│
├── app.py                    # Streamlit web application
├── requirements.txt          # Python dependencies
├── README.md                 # This file
│
├── src/                      # Core analysis modules
│   ├── ai_formulary_analyzer.py
│   ├── formulary_delta_processor.py
│   ├── claude_api_integration.py
│   ├── batch_processor.py
│   └── generate_sql.py
│
├── web/                      # Alternative web interface
│   └── index.html
│
├── docs/                     # Documentation
│   ├── DEPLOYMENT_GUIDE.md
│   └── ARCHITECTURE.md
│
└── examples/                 # Sample data
    └── sample_output/

Pharmacy Operations

• Identify drugs requiring new prior authorization
• Track quantity limit changes
• Monitor formulary additions/deletions

Clinical Teams

• Detect therapeutic substitutions
• Review specialty designation changes
• Assess patient impact of restrictions

Finance Teams

• Calculate cost impact of tier changes
• Estimate copay shift to members
• Project administrative burden

IT/Data Teams

• Generate SQL for incremental database updates
• Export change data for downstream systems
• Automate monthly processing workflows

• Python 3.8+ - Core language
• Streamlit - Interactive web interface
• Pandas - Data processing and analysis
• Claude AI - Intelligent insights (optional)

• Processes 500,000+ records in ~30 seconds
• Handles files up to 100MB+
• Tested with CMS PUF formulary data (57MB files)
• Efficient delta detection using set operations

1. Push your code to GitHub
2. Visit share.streamlit.io
3. Connect your repository
4. Click "Deploy"

Your app will be live at: https://your-app.streamlit.app

Full deployment guide →

• Deployment Guide - Step-by-step deployment to Streamlit Cloud
• Architecture Overview - Technical design details
• User Guide - How to use the platform
• API Reference - Integration documentation

Contributions welcome! Please open an issue or submit a pull request.

MIT License - see LICENSE for details.

**

Built to solve real healthcare data challenges. Questions? Open an issue!