
Fight Medicare Fraud With AI, Not a Sledgehammer
CMS & the industry should use AI to better target anti-fraud efforts
There is an old saying that when the only tool you have is a hammer, every problem starts to look like a nail. I’ve even heard it applied to Medicare and Medicaid fraud.
So, it’s probably no surprise that as a self-proclaimed fan of AI and original member of the Technology Digirati, I can’t help but see another tool we should be using.
Besides, healthcare already has plenty of hammers. I’m not sure enrollment moratoriums need to be one of them.
Enter AI.
The Centers for Medicare & Medicaid Services (CMS) allowed its Medicare enrollment moratorium affecting certain durable medical equipment (DME) suppliers to expire. The response from the DME side of the industry was interesting. It wasn’t, “Great, now leave us alone.” In fact, it was almost the opposite.
The message was: Keep fighting fraud, just get better at it.
Narrowing the Target
That distinction matters, particularly for those in home health and hospice (where CMS’ enrollment moratoria have recently been extended another six months).
Nobody wants fraudulent providers entering Medicare. Nobody wants taxpayer dollars going to sham operations. And nobody wants bad actors damaging the reputation of an entire industry. But I’m also not convinced that stopping everyone from entering a market is the best long-term answer.
It works, certainly. If nobody can get through the gates, the bad guys can’t either.
Of course, neither can the good guys.
And in this day and age, we should be capable of something a little more sophisticated.
We now have AI, advanced data analytics, electronic prescribing, electronic verification, identity verification and enormous amounts of healthcare data. We should be asking how those tools can help us distinguish legitimate providers from organizations that deserve a much closer look.
Home health and hospice seem like particularly good places to start.
Think about the information already available: ownership, leadership, referral patterns, physician orders, staffing, geography, billing, utilization, changes in volume, relationships between individuals and other healthcare organizations.
One of those things looking unusual probably doesn’t mean much. Five or six of them looking unusual at the same time—now I might want somebody to take a look.
That’s where I think AI becomes particularly interesting. Not because AI should decide who gets into Medicare (please don’t give the robot that job!), but because it can help identify patterns that would be difficult for a person to see across millions of pieces of information.
That allows the human beings who actually understand healthcare, fraud and compliance to spend more of their time looking at the organizations that warrant attention.
And that is the larger opportunity.
We don’t need less oversight. We need smarter oversight.
Providers & Accreditors on Board
I would actually argue that legitimate providers should welcome it. If you’ve invested in compliance, qualified staff, appropriate leadership and quality patient care, a regulatory system that gets better at distinguishing you from a fraudulent operator is, as Martha Stewart would say, “a good thing.”
Accreditation can play a role here as well.
Accreditors see healthcare organizations differently than claims systems do. We see operations. We see policies put into practice. We see corrective actions, quality processes and whether an organization actually functions the way it says it does.
There should be opportunities to think about how that information, appropriately protected and responsibly used, contributes to a better picture of risk.
None of this means enrollment moratoriums should disappear. There may absolutely be situations where CMS needs to pump the brakes while it deals with a serious problem.
But we shouldn’t have to deploy the airbags every time we see trouble ahead. Airbags are there for emergencies, and deploying them comes with consequences for everyone in the car. The better strategy is avoiding the crash in the first place.
DME has already lived through many of these challenges. Home health and hospice can learn from that experience rather than repeat it.
Technology gives us an opportunity to make it increasingly difficult for fraudulent organizations to enter and exploit Medicare without making it unnecessarily difficult for legitimate providers to serve patients.
Use the technology to find the patterns. Use the data to identify the risk. Then put experienced people in a position to make better decisions.
Fight fraud aggressively.
Keep the sledgehammer around for emergencies.
But maybe it’s time to let AI pump the brakes?
OUR DIGITAL PARTNERS






