Surprise new element in DOGE’s Medicaid experiment: taxpayers vs. health care fraud

Terry Gerton You’ve done some looking into a data set that the DOGE team inside the Health and Human Services agency released. It described it as the largest Medicaid data set in department history. From your legal perspective, what was different about this release compared to prior transparency or fraud prevention efforts?

John Barry Well, there’s a couple things. This is the first time that the Department of Health and Human Services has ever released provider-level Medicaid data to the public. Previously, there were very limited ways to access Medicaid provider-level data, meaning data that you could look at specific providers and what they were billing, but that was required often institutional access and cost money. This is the first time that this has ever been available to the public, and specifically, a data set released to the public for the public to look at and try to find issues on their own.

Terry Gerton So this is a bit like crowdsourcing research.

John Barry That’s right, that’s right. There have been some cuts, the federal level, to the Department of Justice during Trump’s administration. Some estimates have it upwards of 4,000 or 5,000 jobs. And that seems antithetical to the government’s push to rout out fraud at the federal level. And so, interestingly enough, to a certain extent, the government has said, you know what, we have less people in the Department of Justice, but we do have a lot of people, and this is the people’s money. I think the message from the Trump administration has been, this is not the government’s money that’s being potentially misused or stolen. This is your money. This is the taxpayers’ money. And so you, the taxpayers, have an incentive to look into this on your own. And there have been some very public cases, particularly in Minnesota and California, that have been highlighted by Trump and JD Vance, and to a certain extent, the Department of Treasury, that have kind of given the federal government some backing to say, hey, you, individual person can really make a difference and help rout out potentially billions of dollars of fraud, and we want you to do that.

Terry Gerton The federal and state agencies already have specialized fraud units to identify improper billing. How does opening this kind of data to the public change that fraud detection model, and what risks might it introduce?

John Barry  I think that this data set presents opportunities, and as you mentioned, significant risks. The opportunities are that you have private — what I’ll call sleuths, or even data scientists, particularly with the advent of AI and data analytics and the access that individuals have to that, that you almost kind of make it a little bit of a game for these people to try to rout out fraud and find what they can and share that with the public. The issue is that data on its own is a starting point. It is not the end point. It does not tell you the whole story. And I think that there is a risk that certain people publicizing data that they may have found, and think that it is clearly fraud, may really not be fraud at the end of the day. And when you have the Department of Justice and the OIG within the Health and Human Services and the FBI and the Financial Crimes Network within the Department of Treasury, those people live and breathe that and are more specialized and have training on really knowing where to spend their time and what to do with data. And so the risk is that there could be some wild goose chases going on, particularly when it’s referred from the public to the state and federal agencies. You do have that risk that maybe some institutions — even just reputational damage from being publicized on social media networks that, hey, this provider seems to have increased their billing 50% over the last four years. Well, that alone is not indicative of fraud. And I think there is a significant risk that people may be getting a bit ahead of themselves in calling something fraud when it isn’t.

Terry Gerton I’m speaking with John Barry. He’s senior counsel at Epstein Becker Green. Let’s pull on that thread a little bit, because the False Claims Act is invoked here, and whistleblower incentives to try to get people to make those claims and also perhaps charge providers under the False Claims Act. How is all of that coming together with a consideration of due process?

John Barry As we had discussed, there is always the chance for reputational damage, but with respect to due process, even a qui tam action, which is a claim brought in the name of the federal government, but by a private actor, alleging that fraud was perpetrated on the federal government. And in those actions, the individual bringing the claim, which is called a relator, has the ability to potentially get part of the settlement if fraud is ultimately found. That does not ultimately do away with the due process requirements of the legal system. There’s still an accusation; unfortunately in our legal system, sometimes when accusations get out there in the public space, that is almost as bad as being convicted. But in the court system, an accusation alone does not mean that somebody is going to be convicted. And bringing a qui tam case alone on your own … the way that most qui tam cases go is that an individual relator will bring a case to the government and say, hey, I think there is fraud, I have this special information, I’m going to bring this case, would you like to participate with me? And often in the strongest cases, the federal government will participate with them. If the federal government does not participate, it often means that the federal government has decided there is not a strong case here or there is not fraud. And so, often those cases will be ultimately thrown out of court by the judicial system saying, well, the federal government didn’t find anything, we the court are going to look at it, did you meet the requirements to really bring a claim here? So, to clarify, the due process requirements are still there with the qui tam case. But you are right that once you bring a claim and once negative publicity gets out about a provider, that is hard to undo that reputational damage.

Terry Gerton I want to explore another angle here. The president has stood up an anti-fraud task force. It’s being led by the vice president. Do you think, or do you see a way where this DOGE HHS data set might be useful and informative in that task force’s efforts? They’re starting with benefits. Medicaid is a great place to start, I suppose.

John Barry That’s right. You know, Medicare and Medicaid spending combined is the largest portion of the annual federal government spending. Those two buckets together account for 27% of annual spending. And so where there is a significant amount of government spending, there will be a significant amount of bad actors trying to perpetrate fraud to get some of those funds. And the Trump administration has been trying to attack fraud in the health care space from as many angles as it can. Not only has it tried to crowdsource this through releasing this HHS Medicaid data, they separately have a whistleblower website set up through the Department of Treasury to specifically incentivize people with financial benefits through reporting fraud that may be tied to health care, but specifically money flowing through banks that would be in the purview of the Department of Treasury. There’s also, in addition to the new task force that was just recently announced, in 2025 there were these health care-fraud data fusion centers that were multi-agency task forces made up from members of the Department of Justice, Health and Human Services, the FBI, to try to gather resources from various government agencies to rout out fraud. So this is all a way of saying the release of this DOGE HHS data set absolutely aligns with the government’s continued push. Almost seemingly every month, there is a new task force or a new release of information. For example, on March 30, the Department of the Treasury, the Financial Crimes Enforcement Network, put out a financial crimes enforcement bulletin specifically about health care fraud and what banks should do, the types of suspicious activity that banks should look for specifically related to health care fraud. So the government is trying to attack this from every angle. And it makes sense because of how large the spending is for Medicare and Medicaid.

Terry Gerton So does this mark, then, sort of a new front in the anti-fraud war, and might we be expecting to see more data set releases, perhaps around student loans or grant recipients or Social Security payments?

John Barry You know, that’s a very interesting question. I think what you see here is this data set is focused on provider fraud. But we may see a movement towards data sets — and this would be very controversial — towards individual or beneficiary fraud. Now the big, beautiful bill tries to get out Medicaid fraud from a different angle, from the beneficiary angle. And the way that the Trump administration is trying to rout out beneficiary fraud is trying to limit the ways in which people may enroll or continue to be enrolled in the Medicaid ranks. For example, there’s the work for Medicaid that someone has to either prove that they’ve been working or going to job training or school for X amount of hours per month. And this is all a way of trying to cut down, in any way possible, on spending in the health care realm. And I think it’s too early to say whether the government would release specific data sets on individuals related to, for example, federal loans. But even before this HHS data set, this is not the first time that the government has made publicly available information about a government program. For example, one of the largest COVID-era spending programs was the Paycheck Protection Program, which were loans to small businesses, and then eventually to even single independent contractors, to try to keep those businesses afloat during COVID. Eventually, every single individual or small business that took a PPP loan, that information was released by the federal government. And even before this HHS data set, people were looking at that Paycheck Protection Program data set to see, hey, should this business really have gotten this loan? That was not publicized. The release of that data set was not publicized in the crowdsourcing way that this HSS data set was. But, you know, this is not the first time that this type of crowdsourcing has ultimately come about through release of federal data sets.

Terry Gerton Coming back to the HHS data, if you’re a provider, what should you be thinking about right now, both for protecting your own reputation and making sure that your data is correct?

John Barry Well, number one, I would try to take a look at that dataset to see whether, if you are a Medicaid provider, how your information has ultimately been relayed through this dataset. One issue is this Medicaid data is coming from the 50 states separately, in addition to certain territories such as Puerto Rico. And so there is always going to be differences in the way that the states report the information. If I were a provider, I would want to look to make sure that the information released publicly at least generally aligns with what I believe that I am billing for, that the state that I’m in or the states that I’m in, are reporting that information accurately. And second, regardless of whether the information is found in the HHS data set or not, providers should always be ensuring that they have a billing and coding compliance program, that they are accurately and properly billing only for the services that they provide.