Related Links

The Cleveland-based payer, with 1.4 million covered lives, has identified many fraudulent claims before they flared into major losses.

"In the old days, if we saw some 'smoke,' by the time investigators got involved there was a fire," says John Shoemaker, the payer's financial investigations unit manager. "By the time we were ready to fight it we already had lost a lot of money."

By some estimates, health care insurance fraud accounts for nearly 3% of the nation's $1.7 trillion in annual health care expenditures. Many payer organizations are using I.T. to help identify claims fraud upfront, when claims are received, or after they are paid, by scanning large amounts of data in search of questionable filing patterns.

Since deploying STARSentinel software from ViPS Inc., Baltimore, Medical Mutual of Ohio can identify potential problems more quickly and minimize losses, Shoemaker explains. ViPS is a subsidiary of Emdeon Corp., Elmwood Park, N.J.

Shoemaker's anti-fraud team includes clinical, financial and law enforcement experts, some with I.T. experience. They investigate some 120 cases of possible fraud each year. Since 2003, they have recovered almost $14 million in fraudulent claims.

Software has played a big role, he notes. Before the days of electronic claims and monitoring tools, Medical Mutual relied on tips-about 400 per month-from various sources, including patients and physicians.

That wasn't enough, Shoemaker says. "We needed an early warning system that looks at claims payments and histories on a more current basis to identify provider claims patterns."

The STARSentinel software is one of the few applications Shoemaker has purchased off the shelf. Its easy customization, which enabled staff to build alerts based on payment patterns over time, has helped Medical Mutual spot problems upfront, correct provider actions and recover payments, he adds.

In one recent example, the system flagged a physician who had billed more than 200 patients as new, "which was odd considering the doctor had been in practice for 10 years," Shoemaker says.

Closer examination revealed that the physician had left one practice and started another and was wrongly billing all patients as new ones. The amount of overpayments recovered wasn't large-about $6,000-but "it stopped a problem that could have gotten worse."

Testing patterns

Medical Mutual also uses the application to run quarterly reports to examine claims patterns. Such runs can unearth duplicate billing or "doctor-shopping" by patients looking to fill prescriptions for narcotics.

Shoemaker also employs data mining software called PolyAnalyst from Megaputer Intelligence Inc., Bloomington, Ind. The software looks at large databases, like the one built and maintained by Shoemaker and his staff, to uncover "patterns and relationships we didn't know existed. Then we can build rules based on those correlations," he explains.

Shoemaker notes there are many rules-based software products on the market. But rules-based analysis "means that I know something. But I need to know what I don't know," such as if there are groups of physicians referring to the same set of providers.

Payer organizations increasingly will rely on I.T. to help them manage the flow of electronic claims, says Louis Saccoccio, executive director at the National Health Care Anti-Fraud Association, Washington. NHCAA has about 100 payer organization members, including Medical Mutual of Ohio, and about 30 law enforcement groups, such as the FBI and the Department of Justice.

The association recently completed an anti-fraud management survey of its payer members that showed 70% are using some type of software-based fraud detection systems, either purchased or internally developed.

While NHCAA members are typically larger organizations, all payers need I.T. tools to keep up. "It's absolutely critical," Saccoccio says. "Most claims are processed electronically and must be processed rapidly because of state prompt-pay laws."