Frequent testing of risk models strengthens Medicare fraud prevention systems, official says

Share this content:

Increasingly sophisticated fraud prevention systems are allowing the federal government to zero in on suspect billing in long-term care facilities, an official said Tuesday.

The Centers for Medicare & Medicaid Services' predictive modeling system reviews about 4.5 million Medicare fee-for-service claims per day, according to David Nelson, director of CMS's Data Analytics and Control Group. In remarks delivered Tuesday at the Predictive Analytics World Government conference in Washington. D.C., Nelson said the agency makes constant tweaks to its risk models to detect possibly fraudulent billing patterns, the Bureau of National Affairs reported.

Nelson explained that every claim received from a provider, such as a nursing home, receives a risk score. Then, claims the claims are sorted and those with the highest risk scores are to Zone Program Integrity Contractors (ZPICs) for further investigation.

The authority of ZPICs is increasingly becoming a headache for skilled nursing facilities.