Senators want better assessment of Medicare fraud detection program

Medicare officials should better evaluate whether a new system designed to spot fraudulent claims and roll back the program's roughly $50 billion in annual improper payments is living up to its potential, a bipartisan group of senators said Tuesday.

The Centers for Medicare and Medicaid Service responded that it had been measuring results from the $100 million system since soon after its July launch and would update Congress on the results soon.

"The predictive modeling program is well on its way to achieving the kind of results CMS expected," agency officials said.

CMS plans to launch the new system nationwide this summer. But without clear metrics to gauge successes and failures neither the agency nor Congress will be able to determine whether the program is doing what it's supposed to, the senators said in a letter to Peter Budetti, director of CMS' Center for Program Integrity.

"As is often said, one cannot manage what one cannot measure," Sens. Tom Carper, D-Del.; Scott Brown, R-Mass.; and Tom Coburn, R-Okla.; said. Carper is chairman of the Senate Homeland Security Subcommittee on Federal Financial Management, which oversees CMS' financial issues. Brown is that panel's ranking Republican and Coburn, a physician, is a subcommittee member.

The predictive analysis tool was designed to flag common patterns of Medicare fraud such as suspicious billing patterns or a great distance between the hospital where treatment occurred and the claimant's home address. The plan is for CMS officials to halt those payments and immediately investigate them for possible fraud.

Legislative requirements that most Medicare claims be paid within 30 days traditionally has meant that CMS paid out claims before investigating them -- what officials call the "pay and chase" model.

The senators' letter concludes with a list of 10 questions focused on how much money the tool has clawed back, what lessons CMS officials have learned from the system's implementation so far, and whether those lessons have changed how the agency pays claims and which Medicare services providers it contracts with.

The predictive analytics program is similar to several fraud detection tools used by the Recovery Accountability and Transparency Board, which tracks spending on President Obama's $840 billion stimulus bill. That program has kept stimulus money lost to fraud at below 1 percent compared with a rate of up to 7 percent for government spending generally.

Plans to roll out similar tools on a governmentwide basis are working their way through Congress and the White House.