Existing Data Could Uncover Medicaid Fraud—If Anyone Uses It
The data to find Medicaid fraud – Charges are brought, hearings convene, and the statistics grow larger. This cycle repeats itself predictably whenever healthcare fraud surfaces in the news. Yet beneath these announcements lies an unchanging weakness in how we detect deception. The information needed to identify fraudulent activities already exists within public databases; the problem is simply that nobody is looking at it properly.
Last month, federal authorities unveiled their 2026 National Health Care Fraud Takedown initiative. The results were substantial: 455 individuals faced charges connected to $6.5 billion in suspected false claims across 56 different federal districts. This figure represented nearly a forty percent increase compared to the previous year’s totals.
Just days later, the House Energy and Commerce Subcommittee on Oversight and Investigations held a hearing focused directly on this persistent issue. State Medicaid directors from California, Minnesota, New York, and Ohio provided testimony about their experiences managing fraud within their respective programs.
A Methodological Approach to Detection
Earlier this year, my research team undertook a field investigation examining home healthcare agencies located in Columbus, Ohio. We began our analysis using publicly available information: provider-level reimbursement records accessible through the Department of Health and Human Services Open Data Platform, along with corporate registrations and licensing documentation.
One particular agency caught our attention. Since 2018, it had collected approximately $11.1 million in Medicaid payments. Despite this substantial revenue, its website featured generic stock photographs rather than images of actual staff or facilities. Its contact email utilized a Gmail address rather than a domain-specific corporate email. When we placed a telephone call, a single word—”Hello”—was all we received. No business name was mentioned, and no proper greeting followed. We explained that we sought a caregiver for an elderly family member, and the person on the line stated that no caregivers were available before hanging up. A subsequent in-person visit revealed an office containing cubicles but lacking any formal intake process or available caregivers. We left our contact information, yet no one returned our call.
None of these observations alone prove fraudulent activity definitively. Home healthcare represents a sector where small operations remain perfectly legitimate, staffing shortages occur genuinely, and utilizing a Gmail address constitutes nothing more than a convenience. However, each observation generates a signal, and our investigation demonstrated a methodological point: this signal appeared within public data long before we traveled to Columbus.
Connecting Billing Volume to Operational Reality
This connection proves crucial because a mid-sized home healthcare organization billing several million dollars annually in Medicaid reimbursements typically maintains between 150 and 250 active patients. Such an operation employs approximately 12 to 18 administrative personnel and occupies a commercial office space ranging from 2,500 to 4,000 square feet. This configuration reflects the practical requirements of coordinating hundreds of caregivers, processing thousands of billable hours each week, and managing payroll, compliance obligations, and billing functions simultaneously.
Billing volume essentially serves as a proxy for operational scale. When these two elements clearly diverge, that divergence becomes observable within data that already exists in public repositories.
The House Oversight Task Force investigating Ohio Medicaid waiver fraud estimated that losses within Ohio’s personal care services program alone reached $1.2 billion. Additionally, a March 2025 Inspector General report revealed that thirty-six percent of all convictions reported by state Medicaid Fraud Control Units during fiscal year 2024 involved personal care services—more than any other program category.
This represents the dominant fraud category within Medicaid enforcement, concentrated precisely in the sector where providers’ physical footprints prove most difficult to verify and where billing volume most clearly indicates what operational reality should resemble.
A Simple Solution Already Within Reach
The deficiency at the heart of this problem proves surprisingly straightforward. The systems responsible for enrolling and compensating Medicaid providers fail to routinely ask whether a provider’s operational reality aligns with its billing profile. They ought to.
Current enrollment procedures examine licenses, verify provider numbers, and confirm tax identification numbers, yet they do not systematically question whether the address associated with a $5 million annual billing operation appears consistent with that level of activity. This question remains answerable using public data, requiring nothing more than an analytical framework that treats billing volume as a proxy for operational scale. Such an approach can identify cases where the two elements diverge, prompting closer examination.
Florida has already progressed in this direction recently. Governor Ron DeSantis announced a Medicaid integrity initiative incorporating enhanced provider screening, an enrollment moratorium targeting high-risk provider categories, and a comprehensive statewide revalidation of all active Medicaid providers.
The critical question now concerns whether other states will adopt similar measures before the next wave of criminal charges gets filed.
The fraud targeting Medicaid home healthcare demands no particular sophistication. It requires merely a registered address, a provider number, and a system that processes paperwork without verifying whether any of the filings reflect genuine operational reality.
