Addressing Fraud, Waste, and Abuse: How Artificial Intelligence Helps Control Costs and Strengthen Healthcare Integrity

Buried within vast seas of data and reams of medical records are subtle anomalies, patterns, and outliers — digital needles in a multitrillion-dollar haystack. Together, they cost the United States healthcare system from tens to hundreds of billions of dollars each year.

Uncovering these subtle signals that something may be amiss can help expose instances of fraud, waste, and abuse (FWA) — hidden inefficiencies, errors, and illicit activities that contribute to rising healthcare costs and point to potentially dangerous and compromised care.

As a partner in health for the people we serve, we strive to control healthcare costs for our members and clients by using traditional techniques and visionary approaches fueled by digital advances. While uncovering and addressing FWA requires time and expertise, Elevance Health associates taking on these challenges have a powerful ally at their disposal: artificial intelligence (AI). 
 

What Is Fraud, Waste, and Abuse in Healthcare?

Though often grouped together, each component of fraud, waste, and abuse carries a specific meaning within the context of healthcare and health insurance. Each FWA element also can take various forms. Broadly speaking:

• Fraud can occur when someone intentionally deceives a health insurer — including private companies and government programs like Medicaid and Medicare — to obtain money or benefits.
• Waste often involves the overuse or misuse of healthcare resources or services, resulting in unnecessary costs. It’s typically considered unintentional, as opposed to intentional fraud.
• Abuse can occur when standard medical or billing practices aren’t followed, leading to unnecessary costs or inappropriate care.

Where AI Insights Meet Human Oversight

We work to prevent FWA and its drain on the healthcare system in part by assessing claims and billing patterns for unusual or suspicious charges. Increasingly, this work pairs the human expertise of Elevance Health associates — primarily within the Payment Integrity division of our Carelon healthcare services business — with AI-driven lead identification or predictive AI.

The process often begins with models that analyze claims and payment data to ensure services have been billed and paid for appropriately. These analyses can highlight unusual patterns: Perhaps a cardiologist in one county has used different coding when submitting billing claims for echocardiogram procedures than his peers in surrounding counties, for example.

A human investigator can look further into this to determine whether the variations may simply have been a mistake; whether the discrepancies are due to different network, contract, or legal regulations; or whether the provider might be intentionally upcoding — using billing codes attached to a higher level of service or more complex diagnosis than what was delivered.

The growing use of AI-supported tools across the healthcare system — including among care providers — makes advanced analytics even more important for identifying unusual patterns and potential problems.

“Advanced analytics and data-driven insights help guide our investigative strategy,” said Beth Franke, staff vice president of our Special Investigations Unit. “When indicators suggest elevated risk, our team conducts a focused review to determine if fraud, waste, or abuse is occurring.”
 

Data Tools That Drive Efficiency and Control Costs

Many divisions across Elevance Health contribute to our FWA prevention efforts. Areas such as information security, care provider credentialing, local health plan operations, and information technology all play a role in protecting the integrity of our programs, which aids in controlling costs.

In the payment integrity area, teams specializing in tasks like auditing, investigations, data mining, and benefits coordination have various AI-powered tools available to help them ensure proper billing and payment and curb fraud, waste, and abuse. These programs can summarize thousands of pages of medical records, for example, or align formatting for care provider contract data and medical claims to support any required human review of irregular payment patterns.

By integrating AI, our teams can work to address issues earlier in the claims process, and they can perform tasks — like time-consuming audits — significantly faster. Within a clinical auditing unit in Carelon’s Payment Integrity division, for example, one program has helped cut associates’ medical record review time in half — from an average of 40 minutes to 20 minutes — across an annualized volume of nearly 23,000 reviews.

When an issue is discovered, our associates often start with education as a remedy. This may involve first informing a care provider that they’ve been using the wrong billing code for a procedure or that they’ve prescribed a medication only covered by a member’s benefits under certain circumstances. Such steps help keep costs down for members and protect overall health system integrity by ensuring healthcare dollars are used in accordance with evidence-based guidelines and benefit design. This helps to limit avoidable cost pressures that could affect consumers in the future.

Our associates also can take steps that include trying to recover payments made improperly, recommending contract terminations, or pursuing other actions.

“AI gives us the ability to see patterns humans might not on their own, but it’s our people who bring context and clarity,” said Matt Glynos, vice president of Carelon Payment Integrity. “Equipping our associates with these tools helps us protect our members, ensure integrity in care, and reduce healthcare costs overall.”