With many states admitting to substantial fraudulent payments due to poor management and problematic unemployment insurance (UI) application systems, what are they doing about fraud in other state programs that collectively make up another giant elephant in the room?

While California alone has admitted to processing over $11 billion in fraudulent UI claims in the last year, its monstrous Medicaid program (known as Medi-Cal in California) may have been experiencing double that number of losses for decades.

California’s Medi-Cal program administered by the California Department of Health Care Services (DHCS) provides free or low-cost health coverage for children and adults with limited income and resources – and that amounts to a whopping 30 percent of the Golden State’s 40 million residents. With an overall annual budget now over $120 billion, Medi-Cal is second only to education among major categories in the state’s total spending.

Medicaid Fraud Widespread

Regardless of the recent UI troubles nationwide, across the 50 states Medicaid programs are the state budget item most often cited for fraud.

The U.S. Government Accountability Office (GAO) estimates the Medicaid-related fraud losses nationally at eight percent, and other estimates go a lot higher. For example, Harvard University’s Malcolm Sparrow in his eye-opening tome, “License to Steal,” suggests that improper payments account for 20 percent of spending in Federal/state health care programs.

In California, those estimates mean the state could be losing $10-$24 billion each year to fraud, almost 40 percent of which is state general fund dollars, with the remainder paid by the Federal government. Thus, the need for comprehensive fraud prevention solutions for states is clearly evident. And the advent of new enabling technologies involving data mining, intelligent analytics and artificial intelligence (AI) could not be better timed.

Problems With Management, Bureaucracy

As is often the case with government IT projects however, the problems encountered often are not with the technology, but with the management and the bureaucracy.

For example, in California’s case, the assessment of UI claimants’ identity and eligibility was done without a lookup in the state prison population database – ostensibly to protect prisoners’ privacy, according to government officials. As a result, state officials revealed last month that 31,000 prisoners submitted claims and $400 million in fraudulent unemployment benefit benefits were paid out.

Secondly, agency leadership can also be a challenge.

As state CIO of California, in 1997 I tried to do get the director of the California Department of Health Care Services to institute a new, rigorous anti-fraud program that had been so successful in Massachusetts, saving nearly $2 billion. At the time AI was only science fiction, but there were available data tools and technology solutions that could have saved comparable funds in California. However, the DHCS director abstained, claiming the organization already was doing a good job on fraud issues.

Interestingly, just over 20 years later, DHCS finally did embark on a major Medi-Cal fraud initiative – the Medi-Cal Program Integrity Data Analytics (MPIDA) program. Following the traditional dilatory procurement process of over one year, a $22.5 million MPIDA contract was awarded and work began the following year in 2018.

Now into its fourth year, DHCS officials informed us that the MPIDA project has significantly increased their ability to identify fraud in a more comprehensive manner. DHCS also pointed out several examples of MPIDA success:

• In 2019, MPIDA identified a medical clinic as a risky provider. Through a subsequent investigation, DHCS confirmed the fraudulent provider activity. As a result of the investigation, DHCS decertified the provider, submitted a criminal referral to the Department of Justice, and saved the state of California $1.2 million.
• Also in 2019, MPIDA identified a risky pharmacy through the “cluster/network” analysis. The analysis indicated fraudulent billing behavior and led to an inventory reconciliation. Eventually, the provider was suspended from the Medi-Cal program, and DHCS recovered $1.2 million.
• In 2020, MPIDA identified a pharmacy through multiple fraud flags (dispensing to deceased beneficiaries, patients residing a number of miles away from providers, and outlier billing patterns of antipsychotic medication) that indicated a potentially high-risk provider. As a result of an audit and reconciliation of the pharmacy, DHCS issued a demand to the provider for $7 million and submitted a referral to the California Department of Justice.