Healthcare insurers have been held back by rules-based, inefficient technology that still allows $240 billion in annual fraud, waste and abuse. Industry analyst firm Aite Group compares Mastercard’s revolutionary use of Brighterion AI in other payments fraud and finds the same platform could transform FWA.

When Mastercard analysts started digging into the healthcare fraud, waste and abuse (FWA) problem, they knew they could help. By using (and later acquiring) Brighterion’s AI-based fraud prevention solution, they transformed the payments fraud environment. Why not use AI for FWA prevention as well?

After assembling a team of payments fraud specialists, FWA investigators, data scientists and others, Mastercard Healthcare Solutions was born. Fast forward to Spring 2020 and Aite Group was invited to provide analysis to the newly developed FWA solution.

“Mastercard has revolutionized a highly rules-based, inefficient payment fraud space by leveraging its AI technology in its card network to identify and stop fraudulent card payments on a real-time basis,” Aite reports in Putting AI to Use in Fraud, Waste, and Abuse. “The company is now capitalizing on that experience and applying it to tackling FWA in healthcare payments.”

Predictive analytics prevents pay and chase

Aite analysts cited that fraud is significantly reduced through the solution’s unique combination of AI tools. In particular, Mastercard AI uses unsupervised learning to update schemes in real time while supervised learning reveals patterns and anomalies. The result is high accuracy and continuously improved outcomes.

They also validated our thinking: by implementing early detection, FWA will be dramatically reduced before payments are made to providers. The traditional claims process often ends with paying and later chasing providers for reimbursement of funds paid for incorrect or fraudulent claims. Mastercard’s strategy is early detection to prevent and save incorrectly paid claims at the outset.

“The prepayment phase of the claim workflow is ripe with opportunity for automation,” Aite analysts wrote. “Intervening earlier in the claim process, or preventing and saving, can yield better results than the post-payment model, or the pay and chase model. The post-payment model tends to be costlier and time-consuming due to the manual interventions involved and the time required to build and follow through on a case.”

Learn more about the various claims models, types of fraud and Aite’s analysis by downloading Putting AI to Use in Fraud, Waste, and Abuse.