The Department of Government Efficiency (DOGE) open-sourced on Friday what it describes as the largest Medicaid dataset in the department’s history, containing aggregated provider-level claims data from January 2018 through December 2024.
Dataset Enables Fraud Detection Analysis
The dataset compiles individual claims by provider, procedure, and month, showing the number of beneficiaries served, claims submitted, and total amounts paid by Medicaid.
The data covers outpatient and professional claims with valid HCPCS codes across all states and territories.
The DOGE HHS stated on X that the dataset could enable the detection of fraud patterns, citing the large-scale autism diagnosis fraud seen in Minnesota as an example.
“Using this dataset, it would have been possible to easily detect the large-scale autism diagnosis fraud seen in Minnesota,” the DOGE HHS wrote on X.
The data covers fee-for-service, managed care, and Children’s Health Insurance Program claims.
Dataset Specifications and Privacy Protections
To protect beneficiary privacy, the dataset applies cell suppression, meaning rows with fewer than 12 total claims are dropped entirely.
The accuracy of the data depends on submissions from each state to the Transformed Medicaid Statistical Information System, which has known data quality issues that vary by state and by data element.
Tesla Inc. (NASDAQ:TSLA) CEO Elon Musk, who is also the former chief of DOGE, stated on X, “Medicaid data has been open sourced, so the level of fraud is easy to identify,” adding that “DOGE is not a department, it’s a state of mind.”
Financial Advisor Mark Quann noted on X that the entire U.S. government should be put on open source, listing agencies including Medicare, the Pentagon, Veterans Affairs, and the IRS, arguing that if taxpayers fund it, taxpayers should be able to see it.
This development comes a year after the DOGE team gained access to the Centers for Medicare and Medicaid Services (CMS).
Photo: Shutterstock
Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors.
Recent Comments