When It Comes to Fraud in NEMT Programs, It’s Time to Stop Talkin’ and Start Fixin’

By: Dan Reid, President/Managing Member, Grove Transit

Congress has named non-emergency medical transportation a fraud risk, and a new bill takes aim at it. The bill is a good start, but it borrows a tool built for home care. Here is how to make it work for transportation.

In February, federal and state prosecutors in Colorado announced indictments on two people for defrauding the state's Medicaid NEMT program through various fraudulent acts including "ghost rides" that never happened, many of which were for the provider or her family members, overstating mileage on trips, billing for trips after a beneficiary had died, and billing for trips that were not to medical destinations.   None of these schemes required a great deal of sophistication, in fact, they were made possible because they were allowed to utilize processes that require little sophistication in terms of information verification, and a system that pays first and checks later.

Both cases appear in the House Energy and Commerce Committee's September 29, 2026 Majority Staff Report, 119^th Congress; The Real Costs of Fraud: Examining Program Integrity Requirements and Exposign Fraud Medicare and Medicaid.   At the same time, Rep. Gabe Evans of Colorado is planning to put forward the Strengthening Tools for Oversight to Prevent Medicaid Fraud Act of 2026, the "STOP Medicaid Fraud Act" which would require states to expand Electronic Visit Verification (EVV) to verify NEMT and applied behavior analysis services.

As an NEMT operator in Mississippi and Louisiana I have a stake in this. As a long-time proponent for improving program integrity in the NEMT industry, I want the report to have an impact, and I want the bill to pass - and I want it to work. But, I have concerns that the report serves to highlight issues but is bereft of solutions, and the bill, as drafted, won’t stop most of the schemes the Committee documented.

However, with just a few changes, most of which I have been advocating for with TTA for some time now, these schemes could be stopped before a dollar goes out the door.

The report covers all of Medicare and Medicaid, but NEMT shows up in its headline findings. The Committee lists NEMT among five Medicaid services "experiencing high rates of fraud," alongside applied behavior analysis, home and community-based services, adult day care, and substance use disorder treatment. The NEMT-specific evidence cites not just the Colorado examples above, but other identified issues, and notes that The HHS Office of Inspector General has an active targeted review of fraud, waste, and abuse in Medicaid NEMT.

The Committee's central lesson is consistent with what we, concerned NEMT providers, have been saying: "detect and prevent" beats "pay and chase." As the report puts it, "very few dollars are actually recovered once payments are released."

The STOP Medicaid Fraud Act adds a short new subsection to the Social Security Act. Starting January 1, 2027, states would be required to expand EVV to NEMT and applied behavior analysis. Their EVV system must verify the same six data points already required for home care under the Electronic Visit Verification law: the type of service, the person receiving it, the date, the location of service delivery, the person providing it, and the start and end times.

A state that falls short gets a notice from CMS, has 90 days to submit a corrective action plan, and has one year to come into compliance. States that already require "any system" for electronic verification are exempt, as long as they keep it.

I applaud this meaningful step as too many NEMT programs still allow paper trip logs, handwritten odometer readings, and passenger signatures that verify only that someone signed a line. Requiring electronic records moves the whole industry forward.

However, home care EVV was designed to answer one question: did a caregiver show up at a home at the scheduled time? A ride is a different thing. It has a path, a distance, a destination, a driver, a passenger, and a vehicle. The six home care data points capture almost none of that. Measured against the Committee's own cases, the gaps are clear:

Scheme in the report	The bill as drafted	With the changes below
Ghost rides	Partly. A phone check-in can be faked by a colluding driver or rider.	GPS trip data, telematics, and driver authentication must all agree.
Rides to non-medical destinations	No. Records that a trip happened, not where it ended.	Appointment confirmation ties the trip to a covered service.
Inflated mileage (400+ mile trips)	No. Mileage is not a required data point.	Mileage comes from GPS, not from a handwritten claim.
Billing after a beneficiary's death	No. No eligibility check required.	Eligibility checked at booking and before payment.
Wrong driver or vehicle	No. Vehicles are not covered.	Credentialed driver and vehicle verified at the trip.

Put simply, the bill requires states to record rides electronically. It does not require anyone to check that record before paying the claim. That is just pay and chase with better paperwork.

Three structural issues compound this. The exemption for states with "any system" could cover a state whose system captures very little. The enforcement mechanism, a corrective action plan with a one-year runway, carries no financial consequence. And a January 2027 start date gives states, managed care plans, brokers, and providers about three months to comply, which invites delay and waivers rather than real adoption.

Five simple changes would close many of the gaps left by the proposed legislation.  None of these require revolutionary new technology, and the technology is currently in use by most rideshare apps, delivery services, and commercial fleets, including many NEMT providers.

1. GPS Data Verification.  Require geocoded pickup and drop-off locations, time-stamped arrival and departure events, and mileage calculated from GPS data through the vehicle's telematics, not typed in by a driver or read off an odometer. This single change would flag overstate mileage, trips not performed, and drop-offs at non-medical destinations, not years down the road when an audit is performed but the same day a trip is performed.  States can then eliminate paper-era requirements like odometer logs and passenger signatures, which add burden without adding assurance.

2. Verify Before Paying. Make complete, consistent trip data a condition of payment. Trip records should pass through an automated validation gate before a claim can be submitted, with exception flags for impossible travel times, overlapping trips, mileage anomalies, duplicate billing, and mismatched driver or vehicle information. The beneficiary's eligibility, including date of death, should be checked when the trip is booked and again before payment.

3. Confirm the Destination. Where a medical appointment exists, the entity that authorizes the ride should confirm it electronically through standard health data interfaces (HL7), exchanging only the minimum necessary: that the appointment exists, its date and time window, and the healthcare provider's location. No clinical information needs to change hands. Pharmacy stops and similar trips that are not scheduled in a medical record system can stay under existing verification rules.

4. Verification of Driver and Vehicle at the time of service. The EVV requirement to verify the "individual providing the service" is, in practice, usually a login, and logins can be shared. Driver identity should be confirmed at the time of service with authentication that cannot be handed to someone else, such as the same fingerprint or face recognition used to unlock a smartphone. The vehicle should be identified through its onboard diagnostic port, which ties the trip to a specific VIN. That also shuts down upcoding: a trip billed as wheelchair transport must be performed by a vehicle credentialed as wheelchair accessible.

Behind this should sit a centralized credentialing exchange for each state, drawing on federal and state exclusion lists and linking provider ownership, and driver and vehicle credentialing, so dispatch software simply cannot assign a trip to an excluded, expired, or unqualified driver or vehicle. This puts into practice the Committee's recommendations on stronger provider screening, exclusion checks, and ownership transparency, and is directly on point with verify and prevent, rather than pay and chase.

Make it reasonable and enforceable. Apply the requirement to every delivery model: fee-for-service, managed care, and broker programs. There should be an open data standard that states and brokers must accept from any compliant system, so providers are not forced onto a single state-selected vendor and states get consistent, auditable data. Current state systems should be exempt if they meet the full standard, not just “any system” in place. Force a renewed focus on program integrity by holding the paying entity financially responsible when it pays a claim with incomplete verification data. And finally, replace the January 2027 date with a phased schedule, starting with digital capture and pre-payment validation, then adding appointment confirmation and point-of-service authentication, and tie continued noncompliance to federal matching funds.

There is a second benefit. A large share of improper NEMT payments come from missing or inconsistent documentation on trips that really happened, rather than intentional fraud schemes. These same controls would minimize the chances of inadvertent false or incomplete claims being submitted, let alone paid.  Investigators wouldn’t have to spend time on these unintentional errors, because they would be prevented up front, and could focus their efforts on the people deliberately stealing from the program.

Honest providers benefit too. Right now, a legitimate company can lose payment on a completed trip over a missing signature, while a fraudulent one can bill for rides that never happened and not be caught for years. Verified trip data protects the first and exposes the second.

The Committee has correctly identified program integrity as an issue in the NEMT industry, and the STOP Medicaid Fraud Act is right to reach for electronic verification. But verification that is only recorded, and not checked before payment, will not stop the schemes in the Committee's own report. And verification that is focused on the wrong metrics won’t verify the trip before paying, confirm the destination, prove who drove, or make the rules uniform. Do that, and fraud in NEMT becomes very hard to commit and very easy to prove.

---

Dan Reid is President of Grove Transit, a NEMTAC-accredited NEMT provider in Mississippi, and Go Ride in Louisiana. He is a Board Member and Immediate Past-President of The Transportation Alliance and serves on the NEMTAC Board of Directors.