Statistical flags indicate unusual patterns — not proof of fraud or wrongdoing. Read our methodology

Updated with 227M records · Feb 2026 HHS Data Release
Last updated: February 2026

Follow the money.
$1.09 trillion in Medicaid spending.

We analyzed every Medicaid billing record released by HHS — and ran code-specific fraud detection across 617,000+ providers. 1860 providers raised red flags.

Total Spending

$1.09T

2018–2024

Billing Records

227.1M

Individual claims

Providers Analyzed

618K

Unique NPIs

Flagged Providers

1860

statistical + ML analysis

Procedure Codes

11K

HCPCS codes

Key Findings

1860

Providers Flagged

Using 13 statistical tests, 5 advanced detection methods, and machine learning trained on 514 confirmed fraud cases. None appear on the OIG exclusion list.

View watchlist

$122.7B

Personal Care Spending

Code T1019 alone accounts for 11% of all Medicaid spending. The OIG calls personal care the #1 fraud-risk category.

See details

0 of 1860

On OIG Exclusion List

Cross-referenced against 82,715 excluded providers. Zero matches — suggesting our analysis surfaces new, uninvestigated activity.

Our methodology
Featured InvestigationCritical Risk · 7 Flags

6,886%

Community Assistance Resources & Extended Services INC grew from $1.6M to $112.6M in one year

A 6,886% spending increase in a single year while billing 4.5× the median rate for skills training services. Flagged by 7 of our 13 statistical fraud tests.

NPI: 1396049987·NY·View full profile →

Spending Growth

Full analysis →

Medicaid spending grew $76.32B from 2018 to 2024

-6.9% YoY

Highest-Spending Providers

View all 1,000 →

Top States by Provider Spending

All 50 states →

Latest Insights

All stories →

How We Did This

We analyzed 227 million Medicaid billing records released by HHS, covering 617,503 providers and 10,881 procedure codes from 2018–2024. We combined 13 statistical fraud tests — including 4 code-specific smart tests that compare each provider's cost per claim against the national median — with a random forest ML model (AUC: 0.77) trained on 514 OIG-excluded providers. These are unified into risk tiers (Critical, High, Elevated, ML Flag) for a single view of the most suspicious billing patterns.

Built with227M HHS Records13 Statistical TestsRandom Forest ML (AUC 0.77)OIG Exclusion List Cross-Ref

1860

Red Flags in $1.09 Trillion

We analyzed every Medicaid provider in the country using 13 statistical tests and machine learning. See who got flagged.

View the Risk Watchlist