Skip to main content
How AI and Machine Learning Can Fight Health Care Fraud

Health care payers use a variety of tools and solutions to fight fraud, waste and abuse in their fee-for-service healthcare claims. Recently, artificial intelligence (AI) and machine learning have become popular concepts in payment integrity. But how can they help reduce your losses to health care fraud, waste and abuse?

First, let’s look at what is meant by AI and machine learning in this space.

One way to think about AI is that it emulates a human who is evaluating claim, provider and member/beneficiary risk for fraud, waste and abuse. Think of AI as an individual who can assess each of your claims, providers and members or beneficiaries for risk of health care fraud, waste and abuse with “always on” reliability, keen insight, and consistent results. And AI is programmed to look for suspicious behavior and anomalies – things that don’t look right.

Machine learning learns non-human programmed insights by analyzing data and discovering patterns that a typical human analyst would not.  These insights may be driven by changes in benefits or membership, or changes in claiming behavior in response to new payment integrity controls.

Although rules-based payment integrity solutions play an important role in payment integrity they are only simple AI, as they do not utilize machine learning. AI and machine learning in combination are found in the leading analytics-driven decision management systems for healthcare payment integrity, and are often used in conjunction with systems that use rules.

Putting AI and Machine Learning to Work

How do you use AI and machine learning to best effect in your payment integrity program? We recommend that you consider the following things:
  • Be sure that you’re ready to take the “next step.” Ground your adoption of AI and machine learning in a payment integrity strategy which reflects the specific needs of your organization and applicable “industry best practices,” and which also identifies measurable performance gaps which you plan to address with AI and machine learning.
  • Get your stakeholders on board. Successful adoption of AI and machine learning, like any other change to your payment integrity program, requires the support of your stakeholders. Engage your stakeholders to maximize the payment integrity results you achieve with AI and machine learning.
  • Be prepared for change, because the people who inappropriately take your money will adapt. Make sure that your AI and machine learning can keep up with the bad guys by automatically reacting to changes in claiming behavior.
  • Watch out for hype. There are a range of opinions about what AI and machine learning mean. If you specify AI and machine learning for a payment integrity solution, be clear about what you’re asking for, and ask questions about what a vendor is offering in response to your requirements. Also make sure that their approach includes explainable AI and other techniques which aid in the operationalization of machine learning.
We welcome your comments. For more information, see our white paper on Uncovering More Insurance Fraud with Analytics.

related posts