
Utilizing Human and Artificial Intelligence to Combat Healthcare Scam
CIO Review APAC | Monday, February 27, 2023

The fight against healthcare fraud, waste, and abuse has advanced to the point that it now calls for a combination of human intelligence (HI) and artificial intelligence (AI) skills.
FREMONT, CA: The fight against healthcare fraud, waste, and abuse has evolved to involve both human intelligence(H.I) and artificial intelligence. National Health Care Anti-Fraud Association (NHCAA), delineate that healthcare fraud results in financial losses in the tens of billions of dollars annually. To reduce such losses, healthcare claims must be subjected to a hybrid approach of AI and H.I analysis, which enables the identification of nuanced patterns and the recognition of changing conditions.
A combination of H.I and AI approaches are necessary to combat healthcare fraud, waste, and abuse, which is growing increasingly sophisticated. The use of AI algorithms to compute the data and H.I analysis to assess the claims can enhance the payment integrity review process. The first step to integrating H.I and AI are to identify the issue and gather data, followed by developing a defensible concept and putting it into production. The approach must be updated and improved continuously to remain effective.
Stay ahead of the industry with exclusive feature stories on the top companies, expert insights and the latest news delivered straight to your inbox. Subscribe today.
It is necessary to blend the two approaches into a hybrid approach to detect anomalies. This approach helps to identify an issue and a defensible concept. Anomaly packets and human-identified concerns are sent to the analyst team for review and identification of common patterns/trends. Humans define the defensible concept and, in the case of grey areas, may recommend multiple concepts to be applied in tandem.
The hybrid approach enables rough definitions of "expected behaviour" and the ability to train both analysts and physician teams. Algorithms flag all claims requiring audit by a Special Investigations Unit (SIU) team. Active monitoring is set up to understand if environmental variables have changed. Lastly, new anomalous areas continue to be identified by anomaly packets for review.
A real-life instance of a successful deployment of the H.I/AI approach was an investigation that began with a lab claim for an uncommon genetic immunodeficiency disorder, which led to the identification of testing abuse. Another occasion arose after a provider received a couple of additional claims than expected and ended with the realization that the physician was billing 30 hours a day, 7 days a week.
Despite the high cost of a hybrid approach, the AI and H.I combination can detect suspicious claims, contexts, and subtleties that were previously problematic to identify. A feature-rich space for analysis is essential to this approach since domain experts can tell where to look in the data and what to think about, which helps to point the algorithms in the right direction.
The complementary nature of AI and H.I approach ensures a faster rate of identification of issues, anomalies, and changing conditions. The combination of human expertise and artificial intelligence capabilities provides an opportunity to detect patterns and behaviours that are indicative of fraud, waste, and abuse, leading to better payment integrity outcomes. As healthcare systems continue to evolve and innovate, the hybrid approach will continue to be an essential tool in identifying and preventing fraud, waste, and abuse.
More in News