Advanced machine analysis accurately identifies fraud indicators just two weeks after a claim is filed, much earlier than traditional approaches, insurtech Clara Analytics says.

The US group says its study of 2867 property and casualty claims using the technology “represents a significant advancement in how the insurance industry can approach fraud detection”.

In the study, 9% of open claims were identified as having high potential for investigation referral.  

The tool closely matched actual referrals by adjusters, but it detected possible cases significantly earlier. 

The study also found that awareness of being monitored leads to improved behaviour, and insurers known for effective fraud detection are less likely to be targeted.

“By leveraging advanced analytics, we’ve shown that insurers can identify potential fraud much earlier in the claims process, potentially saving billions in fraudulent payouts,” director of claims solutions Pragatee Dhakal said.

“It doesn’t rely on pre-established fraud indicators. By using unsupervised learning techniques, the system can potentially identify novel patterns of fraudulent activity that might not match historical cases.”

Clara says its technology identifies “cost and treatment outliers” and tracks connections that may indicate fraudulent activity.

“Network analysis revealed important connections between attorneys and medical providers that traditional methods might miss.”

The Federal Bureau of Investigation estimates insurance fraud costs the US industry about $US40 billion ($61.54 billion) a year, excluding medical insurance.