Insurance & Technology is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.


01:07 PM
Russ Schreiber, FICO
Russ Schreiber, FICO

Predictive Analytics and Rules Work Together to Combat Fraud

By using analytics and rules together, insurance claims fraud can be detected at multiple stages for greater effectiveness.

Editor's note: This is part two of a series on combating health insurance claims fraud. To see the first part, which explains how predictive analytics and rules work to identify potentially fraudulent claims, can be read here. This edition focuses on combining the two to create a stronger fraud-fighting organization.

Predictive analytics and rules can have a synergistic relationship when it comes to preventing health insurance claims fraud. As a predictive analytics algorithm identifies emerging but increasingly routine sources of waste, it can help providers develop new rules. The artificial intelligence in the predictive analytics system then “learns” from the new rule patterns and builds increasingly more complex, informed and accurate adaptive models.

In one case study, a pathology group had been adding a professional services claim for analysis on standard blood work. It was just a small amount at a time. While there are many tests that do require analysis, these did not. An analytic system not only identified the discrepancy, it also enabled the payer to implement a rule that no professional services modifiers were allowed with a specific set of tests. Once the rule was created, those claims were automatically caught and no longer appeared in the high-risk claims that were forwarded to analysts, improving operational efficiency and freeing analysts to review new sources of fraud, waste and abuse.

Greater Efficiency and Savings

The analytics-based approach yields dramatic results. Many insurers have seen 20% to 50% reductions in fraud losses and 20% to 25% fewer loss-adjustment expenses. This is why such systems are attractive not only to individual insurers but also to major players in the health care industry. For instance, Emdeon, which operates the single largest financial and administrative information exchange in the U.S. health care system, is combining its Payment Integrity Solutions with predictive analytics powered by FICO to offer analytics-based insurance fraud protection to its client base. (FICO and Emdeon’s have recently published a whitepaper, “Pre-Payment Fraud Detection and its Impact on the Bottom Line,” showing how the best programs meld rules-based software with predictive analytics, accommodating both pre- and post-payment reviews and audits.)

One of the largest health insurers, who processes more than 200 million health, dental, vision, and pharmacy claims a year, uses this combination of predictive analytics and business rules. By adding an additional set of analytic models to score claims, the insurer quickly identified more than 250 new pursuable cases that previously went undetected, including 43 cases that ended up impacting multiple providers.

By using analytics and rules together, insurance claims fraud can be detected at multiple stages for greater effectiveness:

  • Pre-payment scoring: Analytics can identify problems before the check is cut
  • Fast-cycle checkpoints: Analytics can further reduce losses by providing early warnings about the riskiest providers, based on studying a year’s worth of pre-payment scores for each provider
  • Post-payment analysis: Analytics can detect large-scale patterns of fraud and abuse through complex analyses of large batches of claims, detecting patterns of fraud and abuse that would not be evident in smaller data sets

The Bottom Line

These powerful systems flag only the highest-risk claims, reducing the number of legitimate claims delayed for investigation. Providers will know that they can count on prompt and accurate payment, thereby strengthening payer-provider relations.

The right combination of predictive analytics and rules-based analysis insurers to realize pre-payment savings by avoiding unnecessary payments, and post-payment savings by identifying and dealing with suspicious providers. On the other hand, insurance providers will experience systemic savings by correcting policy weaknesses that permitted excessive payments. This three-pronged approach not only reduces costs, but improves customer satisfaction.

Russ Schreiber is VP of the insurance practice at FICO. For more information on FICO’s predictive analytics combating insurance fraud, waste and abuse, visit

Register for Insurance & Technology Newsletters