The Centers for Medicare & Medicaid Services Could Improve Performance Measures Associated With the Fraud Prevention System

The Small Business Jobs Act of 2010 (the Act) requires the Department of Health and Human Services (the Department) to use predictive modeling and other analytics technologies (fraud-detection models) to identify improper Medicare Fee-for-Service claims that providers submit and prevent the payment of such claims. To fulfill this requirement, the Department designated the Centers for Medicare & Medicaid Services (CMS) to develop and implement the Fraud Prevention System (FPS).

OIG certified the actual and projected savings with respect to improper payments prevented and recovered, and the return on investment related to the Department's use of the FPS for each of its first 3 implementation years. However, when performing that work, we became aware that the Department might not have the capability to trace the savings from administrative actions back to the specific FPS model that generated the savings. Without this capability, the Department is not able to accurately evaluate an individual FPS model's performance. Therefore, the Department may be limited in how it assesses the effectiveness of its predictive analytics technologies. We performed this audit to follow up on some of our concerns from our previous audits.

We found that CMS's process for refining and enhancing FPS models needs improvement. Specifically, CMS could not track savings from administrative actions back to the individual FPS models that initiated the investigation because, according to CMS, that capability was not built into the FPS. In addition, CMS did not make use of all pertinent performance results because CMS did not (1) ensure that the adjusted savings Zone Program Integrity Contractors (ZPICs) and Program Safeguard Contractors (PSCs) reported to CMS reflected amounts certified by the OIG and (2) evaluate FPS model performance on the basis of the amounts actually expected to be prevented or recovered (i.e., adjusted savings). As a result, the FPS is not as effective in preventing fraud, waste, and abuse in Medicare as it could be.

As part of its process to redesign the FPS, CMS officials are addressing several limitations. CMS's goal is to transition to a new system that would be more capable of obtaining useful information about FPS model performance and decrease the administrative time and cost that the ZPICs and PSCs spend during their investigations.

Although we acknowledged CMS's efforts to improve the FPS, we recommended that CMS make better use of its performance results to refine and enhance the predictive analytics technologies of the FPS models by ensuring that (1) the redesigned FPS is effective in allowing CMS to track savings from administrative actions back to individual FPS models, (2) ZPICs and PSCs adjust savings reported to CMS to only reflect FPS-related savings amounts, and (3) evaluations of FPS model performance consider not only the identified savings but also the adjusted savings. CMS concurred with our recommendations.