The Centers for Medicare & Medicaid Services (CMS) identified over $210 million in savings during the second year of implementation of its predictive analytics-based Fraud Prevention System (“FPS”), according to a June 2014 report released last week. CMS claimed that these “identified savings” reflected a return of $5 for every dollar spent on the program, a $2 increase over the program’s first year.

But the Department of Health and Human Services Office of the Inspector General (OIG), based on a new metric called “adjusted savings,” calculated a much lower return on investment—only $1.34 for every dollar spent, or $54.2 million. This “adjusted savings” metric uses historical experience in the Medicare program to determine how much of the “identified savings” CMS will actually recover.

The FPS, established in 2010, was designed to prospectively analyze Medicare fee-for-service claims to identify suspicious transactions. Certain triggering criteria are used to identify a claim or a provider as suspicious. For instance, a claim submitted under an ID number reported as stolen would trigger review, as would a physician who billed 100 patient encounters on a single day (as opposed to an average of 30). Claims and providers identified as suspicious are referred to Zone Program Integrity Contractors (ZPICs) for investigation, which can include prepayment claim review, site visits, and other measures.

Since 2011, every Part A and Part B Medicare claim has been run through the FPS. In 2013, the FPS led to 423 prepayment review denials, 235 overpayments being referred to the Medicare Administrative Contractors (MACs), and 75 referrals to law enforcement. But CMS also identifies deterrent effects as an unquantifiable benefit to the program.

The development of the new “adjusted savings” metric, like much of CMS’s report itself, responded to criticisms of the FPS’s self-evaluation metrics levied by OIG and the Government Accountability Office (“GAO”) after the program’s first progress report was released. The oversight agencies had criticized the lack of quantifiable measures used to evaluate the program after the issuance of the first annual FPS report.

This year, both OIG and GAO acknowledged improvements to the program. However, in a report issued as an appendix to the FPS report, OIG still criticized some of the efforts by the FPS and refused to certify $39 million in recoveries that the FPS claimed were attributable to it. OIG stated that FPS lacked sufficient documentation to support its claims. The GAO, in a statement, also called for continuing efforts to improve the prepayment audit system.

In addition to adding the “adjusted savings” metric to its FPS report, CMS also indicated that it had issued new reporting instructions to the ZPICs that would allow them to better analyze whether program savings or administrative actions (such as revocation of enrollment) had resulted from FPS referrals. These changes will allow CMS to directly tie ZPIC actions to overpayments collected by the MACs. OIG called on CMS to issue written instructions to contractors regarding how to implement these changes. The results of these changes should be reflected in future reports.