Objectives

The analysis assists payers in detecting possible fraudulent claims by using data mining and predictive modeling tools. Results summarized in a report can be used by claims analysts in their investigations. Claim types include: Healthcare, Workers’ Comp, and Automobile.

The following cropped screen shot illustrates the identification of three providers submitting very large number of claims for the same patient signaling possible fraudulent.

Analysis and Financial Data

The following indicators and tests are applied to claim data:

• Submission of duplicate claims for the same procedure, same patient at the same time.
• Submission of an unreasonably large number of claims for the same patient over a period of time.
• Submission of an unreasonably large number of claims for many patients over a period of time.
• Submission of claims for the same patient among a number of providers sharing the same patient identity.
• Submission of claims for many different procedures when only one was needed thus increasing the total claim amount.

Benefits

The benefits that can be realized from utilizing these models:

• Reduce the substantial sum of money lost to fraudulent claims by assisting adjusters in identify the most probable and most recoverable instance of fraud.
• Expand fraud discovery by using CRS models that do not rely on the existence of records from previously detected cases of fraud (CRS uses data anomalies and patterns to signal the possibility of fraud).