Helping France's national health insurance provider prevent fraud

By applying sophisticated data analytics techniques HyperCube, with the help of BearingPoint expertise, delivered fraud detection rates of over 70%.

 

One of the key success factors is the collaboration between BearingPoint and the french national health insurance (CNAMTS) on the construction of a database

Pierre Fender, Director of Fraud Control – CNAMTS

Challenge

For CNAMTS, France’s national health insurer, detecting likely fraudulent behaviour before it happens is a key priority. It wanted to identify high-risk profiles and required positive identification rates of 70% or more. Using its own techniques, CNAMTS has so far been able to achieve average positive control rates of between 5% and 40%.

Solution

At HyperCube we analyzed all available and derived data to create a predictive model ready to be rolled out by the client. 

During a four-week sprint cycle as part of the project, we were able to:

  • Build and quality check the database
  • Identify key indicators
  • Create rulesets to pinpoint high-risk profiles
  • Develop a predictive model to detect potential new fraudsters

From this project work we identified three key rulesets that helped to positively identify fraudsters more than 70% of the time. The client has since implemented these rulesets into its own systems. 

Benefits

By implementing the HyperCube rulesets, CNAMTS has a model in place that far outperforms its existing approach. Today, CNAMT is able to:

  • Positively identify fraudsters in over 70% of cases
  • Reduce loss of earnings by more than €40,000 for each fraud avoided

More use cases

Ready for the next generation of analytics?

Try HyperCube and experience its unrivaled range of features and ease of use Get started