Skip to main content

The Intelligent Future of Medicaid Claim Reviews

Medicaid claim review team leveraging AI and machine learning for improved healthcare payment integrity, review claims paperwork in a hospital hallway.

Artificial intelligence (AI) is designed to make life better for humans, but many worry it will have the opposite effect. A survey by Pew Research, for example, shows that people fear being displaced by technology and ultimately surrendering autonomy and control. It’s an understandable reaction, given the depictions of AI in popular culture. Many entertaining — if unrealistic — storylines have centered around sentient computers wreaking havoc while malicious robots rule the world.

It seems that movie directors have yet to seize upon the most exciting and compelling scenario of all: using AI to review Medicaid claims for billing errors.

From Fiction to Fact

AI has already found its way into nearly every industry, including Medicaid. Far from displacing humans, AI is actually benefiting them in multiple ways. For one thing, AI is working alongside humans to make their jobs more efficient. And rather than creating friction, AI can be instrumental in fostering better relationships.

Let’s look at the area of cost containment, where Medicaid programs are charged with being good stewards of public funds. Robust payment integrity initiatives are vital in combating fraud, waste and abuse. Payment integrity programs ensure that claims are medically necessary and correctly submitted, and this is where clinical claim reviews come in.

At their most basic, clinical claim reviews identify claims that were billed or coded in error. The reviews can be conducted in a pre-pay or post-pay environment, and a final determination is performed by an expert clinician.

With the introduction of automation, clinical claim reviews take payment integrity to the next level by adding predictive analytics. Incoming data is analyzed by combining machine learning (ML) technology with historical data for improved accuracy. When Gainwell performs claim reviews, for instance, we rely on more than 10 petabytes of historical data, based on more than 15 years of medical record reviews. When combined with proprietary ML, this data can be used to create algorithms that easily identify the claims most likely to have errors.  

Now, here’s where science fiction becomes science fact. With the use of ML and AI, algorithms constantly improve and get more efficient at targeting erroneous claims. In other words, they not only help to speed the process, but they also continually work to make clinical claim review solutions become even smarter.

The Human Connection

Unlike movie storylines, however, humans are not eliminated in the claim review scenario. In fact, they play extremely important roles.

First, the fine-tuning of algorithms is not accomplished by AI alone. It is actually driven by a synergy between AI and the collective knowledge of data scientists and clinical experts. Without human input, AI can only go so far.

Next, when a human touch is needed, clinical experts can step in to personally review records. These experts can be physicians, nurses, certified coders, dentists or behavioral health professionals, all of whom have the experience and insight necessary to determine whether fraud, waste and abuse are at play.

And here’s one of the best benefits. AI can actually be used to improve human experiences. Used correctly, AI can speed the payment process and actually reduce provider abrasion, saving time, money and frustration. 

Adding Innovation to the Process

When claims can be given an initial review without the associated medical records, a giant stumbling block is removed. The process not only accelerates, but the experience becomes smoother for both payers and providers. The potential for AI to assist is so intriguing that Gainwell chose to test a concept called Intelligent Review as part of our Clinical Claim Reviews solution.

The Intelligent Review process employs predictive models that use data attribute sequencing to analyze combinations of elements within a claim, such as length of stay, diagnosis codes, procedure codes, discharge status and age, as well as provider-specific billing and performance patterns. Algorithms drive the selection process, comparing each sequence to historical outcomes.

Using hierarchal selection, Intelligent Review sends claims with the highest overpayment potential through clinical review without requesting medical records. Each claim is finalized by a clinician before the provider is notified of a preliminary determination. Meanwhile, claims in the next probability tier move through the traditional review process, with medical records requested for clinician review before determination.

Proving a Concept

As part of the test, Gainwell first compared findings from the Intelligent Review process to findings from standard record reviews. After validating the quality of the Intelligent Review algorithms, and determining that they were indeed accurate, we focused on measuring the amount of claims that could flow through this process. In a state recovery audit contractor (RAC) pilot, we found that 25% of claim reviews were able to be conducted without a record request, significantly lowering the administrative burden on providers. And because the Intelligent Review process can be easily deployed in a pre-pay environment, it actually lowers the administrative burden all the way down the line by making sure claims are paid correctly the first time.   

This data-driven, automated approach has now moved from concept to proven solution. Medical records are requested only when required for determination, and providers can still review preliminary findings and submit records for claims they believe were billed accurately, without impacting their appeal rights. For payers, overpayment recovery is accelerated as claims with a high probability of error move into a separate workflow. The review and technical denial process for these claims is reduced from 90 days to 15-20 days. All outcomes, including appeals, continually feed into Gainwell’s AI models, driving future claim selection.  

Embracing the Future

As Intelligent Review has proven, AI can be elevated as an integral part of payment integrity, increasing innovation and helping states make the best use of healthcare dollars. With AI, clinical claim reviews become an even more valuable part of a digital health ecosystem, using smart technology to contain costs and drive innovation in the payment process.

The future of clinical claim reviews is here — and AI is enhancing the process.

About the Author

Gary Call, M.D., is senior vice president and chief medical officer for Gainwell. He collaborates strongly with Gainwell clients and leads our clinical program development and execution. Dr. Call has more than 25 years of experience in the practice of medicine and managed care. Prior to coming to Gainwell, he most recently served as the corporate vice president of clinical programs at Molina Healthcare. Dr. Call graduated from the University of Washington School of Medicine and completed his residency training at the University of Utah. He is a board-certified family physician.

Profile Photo of Gary Call, M.D.