Uncovering Medicaid Fraud
Loading real data...
What We're Presenting
This analysis uses publicly available Medicaid billing data to identify statistical anomalies that may indicate fraud. We've organized and visualized patterns that might otherwise go unnoticed in raw spreadsheets.
- •Provider rankings by billing volume and growth patterns
- •Geographic analysis of fraud hotspots across all 50 states
- •Specialized focus on autism service billing anomalies
- •Interactive visualizations for exploring the data yourself
Where the Data Comes From
This analysis uses publicly available Medicare/Medicaid billing data from the U.S. Department of Health & Human Services (HHS-Official/medicaid-provider-spending dataset).
How We Analyze It
We use advanced data science techniques to identify statistical anomalies that may indicate fraud:
Growth Spike Detection
Identify providers with >200% year-over-year billing increases, flagging unusual growth patterns.
Outlier Analysis
Find providers billing significantly above their state and specialty peers (>3x average).
Pattern Recognition
Detect unusual concentrations in specific service codes, particularly autism-related services (HCPCS 97151-97158).
Technical Stack: Data processing with polars, SQL analysis with DuckDB, interactive visualizations with D3.js.
Taking It Deeper
While this data is publicly available, we've organized and visualized it to make patterns visible that might otherwise go unnoticed in raw spreadsheets.
Each analysis page lets you explore specific fraud patterns in detail—from provider-by-provider rankings to geographic hotspots to specialized service abuse.
"Healthcare fraud costs taxpayers an estimated $30-100 billion annually. Most of it goes undetected not because the data doesn't exist, but because no one is looking at it the right way."
Ready to Explore the Data?
Dive into interactive visualizations, filter by state and growth patterns, and uncover fraud trends hiding in plain sight.
Explore Dashboard →