Highmark Generates $245 Million in Fraud and Waste Savings Using AI

"AI allows Highmark to detect and prevent suspicious activity more quickly, update insurance policies and guidelines, and stay ahead of new schemes and bad actors," says one plan executive.

For 2021, Highmark reported that its Financial Investigations and Provider Review (FIPR) saved $245 million in fraud, waste, and abuse (FWA) prevention. Highmark added this sum to its total FWA savings haul of almost $1 billion since 2017. And while artificial intelligence (AI) may have factored less in Highmark’s efforts in the beginning, it is now central—along with close and diverse partnerships within and outside of the payer’s organization.

The savings breakdown spanned multiple lines of business and included:

• $152 million from employer-sponsored coverage
• $49 million from BlueCard, which provides out-of-network access
• $19 million from Medicare Advantage (MA)
• $16 million from marketplace plans
• $9 million from the Federal Employee Program

Highmark, a member of the Blue Cross Blue Shield Association, operates plans in Pennsylvania, West Virginia, and Delaware—which generated the bulk of the savings—as well as western and northeastern New York.

That the employer-based plans should generate the largest savings is not a surprise. Nor is FWA detection within an extended Blue Cross Blue Shield network. The question is how much of the savings can be attributed to the same providers contracted across programs, with the answer revealing the repeat offenders whose activity may be more intentional than accidental.

Another question is what coding and what does and does not constitute proper practice, particularly in the MA program. As MedPAC has reported: "Documenting additional diagnosis codes increases enrollees' risk scores, which both increases the monthly payment amount a plan receives and increases the rebate amount a plan uses to provide extra benefits to enrollees." Not surprisingly, MA risk scores are continually higher than those of original Medicare fee-for-service claims.

AI provides new tools

FWA is not unique to healthcare. It's not even unique to the claims process. But the growing use of AI to ferret out preventable cost losses is.

"AI allows Highmark to detect and prevent suspicious activity more quickly, update insurance policies and guidelines, and stay ahead of new schemes and bad actors," said Melissa Anderson, EVP and chief risk and compliance officer, in the company's press release.

Highmark provided the following additional detail to HealthLeaders: "Highmark's AI solutions work in concert with our claims adjudication systems and data warehouses. AI algorithms are applied to claims and identify unusual activity within claims. Once identified, the claims and billing providers are reviewed by Highmark's FIPR department. Investigations are opened accordingly to help mitigate future issues and remediate the identified concerns."  

Partnership also key for FWA prevention

According to its press release, Highmark's FIPR partnerships include "health systems, public health officials, law enforcement and other health stakeholders."

Working with these external partners are more than 80 dedicated internal personnel plus numerous Highmark vendors.

The first line of defense in bending healthcare's cost curve is combating FWA. And even as payers and providers seek to strengthen their relationships and collaborate in new ways, identifying fraudulent claims will always part of a health plan's strategy.