Bringing AI To The Fight Against Healthcare Payments Fraud

The coronavirus pandemic poses an unprecedented challenge to healthcare systems that will likely extend for months — and perhaps even years — to come. More than ever, hospitals and other medical facilities need to focus their resources on healing the sick and avoid getting bogged down by administrative costs and inefficiencies.

This brings a longstanding challenge to the fore: Healthcare organizations have long struggled with fraud, waste and abuse (FWA), costing the United States healthcare sector more than $200 billion annually by some estimates. This reflects the complexities of a payment chain in healthcare that involves multiple entities, including patients, insurers, providers and government agencies.

In many realms of businesses, machine learning (ML) and artificial intelligence (AI) have yielded powerful tools to manage such complex matters. Yet, these advanced computational systems have a long way to go in healthcare administration. Just 4.3 percent of organizations in this sector currently use AI, according to PYMNTS’ latest research. In contrast, nearly 10 percent of financial institutions (FIs) use these systems.

This only tells part of the story, however. Healthcare institutions have notably high levels of interest in implementing a specific form of AI that relies on the use of “smart agents,” which can be assigned to multiple entities within a system and offer highly personalized decisioning capabilities. More than a third of healthcare firms are at least “very interested” in implementing smart agent-based AI, while 13 percent are “extremely interested” — four times the share of banks with this level of interest. Moreover, among large healthcare firms with revenues exceeding $500 million a year, a majority (55.6 percent) are very or extremely interested.

These findings and more are revealed in the Unlocking AI Playbook: Healthcare Edition, a collaboration with Brighterion, based on a survey of 47 U.S. healthcare executives that represent a range of institutions with revenues ranging from $50 million to more than $500 million. The study is part of the larger Unlocking AI series by PYMNTS, examining how AI and other computational systems are being used to manage critical business functions, including payments, regulatory compliance, risk assessment and fraud protection.

Another remarkable trend to emerge from our latest study is the degree to which healthcare organizations are interested in deploying smart agent-based AI to improve their anti-fraud systems. Because smart agent-based AI is capable of delivering rapid personalized decisions in transactions involving payers, insurance companies and other entities, it is uniquely suited to address many fraud-related challenges.

Three of the four most important benefits healthcare administrators expect to gain from smart agent-based AI are fraud related, including stopping fraud before it happens, reducing payments fraud and reducing fraud. Moreover, the benefit cited by the greatest proportion of healthcare firms (65.6 percent) is the ability to reduce false positives, or the reducing of blocked legitimate transactions. This underscores the fact that effective fraud detection tools are not only successful in catching suspicious transactions, but also at letting the good ones through.

To learn more about AI’s promising applications in healthcare administration, download the Playbook.

About the Playbook

The Unlocking AI Playbook: Healthcare Edition, powered by Brighterion, seeks to provide a clear and accurate picture of how medical institutions are using artificial intelligence and its potential to optimize their operations.