Skip to main content

Medicare Fraud Prevention: CMS Has Implemented a Predictive Analytics System, but Needs to Define Measures to Determine Its Effectiveness

GAO-13-104 Published: Oct 15, 2012. Publicly Released: Nov 15, 2012.
Jump To:
Skip to Highlights

Highlights

What GAO Found

The Centers for Medicare and Medicaid Services (CMS) implemented its Fraud Prevention System (FPS) in July 2011, as required by the Small Business Jobs Act, and the system is being used by CMS and its program integrity contractors who conduct investigations of potentially fraudulent claims. Specifically, FPS analyzes Medicare claims data using models of fraudulent behavior, which results in automatic alerts on specific claims and providers, which are then prioritized for program integrity analysts to review and investigate as appropriate. However, while the system draws on a host of existing Medicare data sources and has been integrated with existing systems that process claims, it has not yet been integrated with the agency's payment-processing system to allow for the prevention of payments until suspect claims can be determined to be valid. Program officials stated that this functionality has been delayed due to the time required to develop system requirements; they estimated that it will be implemented by January 2013 but had not yet developed reliable schedules for completing this activity.

FPS is intended by program integrity officials to help facilitate the agency's shift from focusing on recovering large amounts of fraudulent payments after they have been made, to taking actions to prevent payments as soon as aberrant billing patterns are identified. Specifically, CMS has directed its program integrity contractors to prioritize alerts generated by the system and to focus on administrative actions--such as revocations of suspect providers' Medicare billing privileges--that can stop payment of fraudulent claims. To this end, the system has been incorporated into the contractors' existing investigative processes. CMS has also taken steps to address challenges contractors initially faced in using FPS, such as shifting priorities, workload challenges, and issues with system functionality.

Program integrity analysts' use of FPS has generally been consistent with key practices for using predictive analytics identified by private insurers and state Medicaid programs. These include using a variety of data sources; collaborating among system developers, investigative staff, and external stakeholders; and publicizing the use of predictive analytics to deter fraud.

CMS has not yet defined or measured quantifiable benefits, or established appropriate performance goals. To ensure that investments in information technology deliver value, agencies should forecast expected financial benefits and measure benefits accrued. In addition, the Office of Management and Budget requires agencies to define performance measures for systems that reflect program goals and to conduct post-implementation reviews to determine whether objectives are being met. However, CMS had not defined an approach for quantifying benefits or measuring the performance of FPS. Further, agency officials had not conducted a post-implementation review to determine whether FPS is effective in supporting efforts to prevent payment of fraudulent claims. Until program officials review the effectiveness of the system based on quantifiable benefits and measurable performance targets, they will not be able to determine the extent to which FPS is enhancing CMS's ability to accomplish the goals of its fraud prevention program.

Why GAO Did This Study

GAO has designated Medicare as a high-risk program, in part because its complexity makes it particularly vulnerable to fraud. CMS, as the agency within the Department of Health and Human Services (HHS) responsible for administering Medicare and reducing fraud, uses a variety of systems that are intended to identity fraudulent payments. To enhance these efforts, the Small Business Jobs Act of 2010 provided funds for and required CMS to implement predictive analytics technologies--automated systems and tools that can help identify fraudulent claims before they are paid. In turn, CMS developed FPS.

GAO was asked to (1) determine the status of the implementation and use of FPS, (2) describe how the agency uses FPS to identify and investigate potentially fraudulent payments, (3) assess how the agency's use of FPS compares to private insurers' and Medicaid programs' practices, and (4) determine the extent to which CMS has defined and measured benefits and performance goals for the system. To do this, GAO reviewed program documentation, held discussions with state Medicaid officials and private insurers, and interviewed CMS officials and contractors.

Recommendations

GAO recommends that CMS develop schedules for completing integration with existing systems, define and report to Congress quantifiable benefits and measurable performance targets and milestones, and conduct a post-implementation review of FPS. In its comments, HHS agreed with and described actions CMS was taking to address the recommendations.

Recommendations for Executive Action

Agency Affected Recommendation Status
Department of Health and Human Services To help ensure that the implementation of FPS is successful in helping the agency meet the goals and objectives of its fraud prevention strategy, the Secretary of HHS should direct the Administrator of CMS to define quantifiable benefits expected as a result of using the system, along with mechanisms for measuring them.
Closed – Implemented
CMS has defined quantifiable benefits expected to result from using FPS, along with mechanisms for measuring them, as GAO recommended. For example, in its 2014 report on the second year's implementation of the system, the agency defined quantifiable measures related to fraud prevention processes. These measures include, among others, the actual amounts of fee-for-service claims payments denied based on edits performed using FPS, and cost avoidance related to revocations of provider billing privileges resulting from the outcomes of the system -- data that can provide information leading to the identification of cases and aberrant provider behavior. Additionally, CMS defined mechanisms for measuring benefits such as return on investment and cost avoidance. For example, return on investment calculations are to consider actual and projected savings compared to the costs of achieving the savings, and calculations of payment amounts denied are to consider adjustments to better reflect actual amounts or payments denied as a result of using the system. By taking these steps, the agency should be better able to determine whether the implementation of FPS is successful in helping CMS meet the goals and objectives of its fraud prevention strategy and to ensure that it continues to provide financial benefits throughout the life of the system.
Department of Health and Human Services To help ensure that the implementation of FPS is successful in helping the agency meet the goals and objectives of its fraud prevention strategy, the Secretary of HHS should direct the Administrator of CMS to describe outcome-based performance targets and milestones that can be measured to gauge improvements to the agency's fraud prevention initiatives attributable to the implementation of FPS.
Closed – Implemented
CMS has described outcome-based performance targets and milestones that can be measured to gauge improvements in the agency's fraud prevention initiatives resulting from the use of FPS, as GAO recommended. Specifically, the Department of Health and Human Services (HHS) defined in its fiscal year 2018 budget justification targets and milestones for measuring the extent to which the use of the system contributed to improvements in the agency's fraud prevention initiatives, such as increases in the percentage of high-risk Medicare providers and suppliers against which CMS takes administrative actions. In the budget justification, HHS identified targets and milestones for measuring increases in actions such as pre-payment claim reviews, revocations of Medicare billing privileges, and suspensions of payments as measures for gauging the performance of the system. As a result its actions, the agency is better able to measure the benefits achieved as a result of using FPS and to determine whether the system is effective in supporting the agency's efforts to improve outcomes of its initiatives to prevent fraud in the Medicare program.
Department of Health and Human Services The Secretary should direct the Administrator of CMS to develop schedules for completing plans to further integrate FPS with the claims payment processing systems that identify all resources and activities needed to complete tasks and that consider risks and obstacles to the program.
Closed – Implemented
CMS finalized schedules and completed plans for integrating FPS with its claims payment processing systems in the second year of implementation. As a result of completing this activity, CMS achieved financial benefits during the second and third implementation years. Specifically, the agency reported, and the HHS Office of Inspector General confirmed, that it achieved $6 million in savings as a result of implementing FPS edits and automatically denying payment of potentially improper claims. By following through on its plans to integrate FPS and other claims processing systems, CMS has demonstrated that the implementation of FPS has been effective toward improving the agency's ability to identify and prevent payment of fraudulent Medicare claims.
Department of Health and Human Services The Secretary should direct the Administrator of CMS to conduct a post-implementation review of the system to determine whether it is effective in providing the expected financial benefits and supporting CMS's efforts to accomplish the goals of its fraud prevention program.
Closed – Implemented
As GAO recommended, CMS conducted a post-implementation review of the Fraud Prevention System (FPS) to determine whether the system is effective in providing expected benefits and supporting efforts to accomplish the goals of the agency's fraud prevention program. Specifically, beginning in 2012, CMS conducted annual assessments of outcomes and return on investment achieved as a result of using the system. Among other things, reports on the results of the assessments identify the number of providers that CMS took action against based on leads generated by FPS and the amount of improper payments of Medicare claims that were avoided as a result of actions taken against providers suspected of submitting fraudulent claims. By assessing the outcomes resulting from the implementation of FPS, the agency is able to determine whether the implementation of the system is effective in providing the financial benefits the agency expected and is successful in supporting efforts to accomplish the goals of its fraud prevention program.

Full Report

GAO Contacts

Topics

MedicareCriminal investigationsMedicare fraudMedicaidSmall businessInformation technologyFee-for-serviceProgram beneficiariesPerformance measurementFraudulent payments