The Centers for Medicare and Medicaid Services (CMS) has implemented a “predictive analytics system” to detect suspicious billing practices in real time by using algorithms and modeling to analyze Medicare fee-for-service (FFS) claims.  All Medicare FFS claims have been streaming through the new system since June 30, 2011.  The system, mandated by Section 4241 of the Small Business Jobs Act of 2010, is similar to those used by credit card companies to detect potential instances of fraud.  It builds profiles of providers, networks, billing patterns, and utilization, which enable CMS to assign “risk scores” to estimate the likelihood of fraud and flag unusual billing activity for more thorough prepayment review.  CMS analysts then review the flagged cases in detail and, if evidence of fraud is found, cases are referred to the CMS Center for Program Integrity, Medicare Administrative Contractors (MACs), and Zone Program Integrity Contractors (ZPICs), as appropriate.

Claims are not being denied solely on the basis of alerts generated by the predictive modeling system.  Risk scores are used by CMS only to prioritize high-risk cases for targeted review by analysts.  Although the risk scores themselves are unappealable, providers’ existing appeal rights are not affected in the event of an overpayment determination.  MLN Matters® Article Number SE1133, which describes the new predictive modeling system, is available by clicking here.