Health insurers fight back against fake claims with fraud-detection software.
About a year and half ago, a Medicare examiner noticed a run on powered wheelchairs around Dallas. Health-insurance payouts for the chairs had soared by 300% in the region, and a quick review showed supersized payouts in parts of California and Florida as well. By the time the inspectors combed through all national data for wheelchair payouts, Medicare found that paid claims had tripled across 21 states in four years--a big anomaly.
The perpetrators got taxpayer-ID numbers and other data, submitted false claims for the $5,000 wheelchairs, and absconded from the country with millions of dollars, after evading investigators for four years, says Kimberly Brandt, a top fraud investigator for Medicare and Medicaid. Slow analysis was partly to blame. "We had seen a lot of little blips," she says. "But we couldn't pull all the data together."
Last September, Medicare slapped new software checks on its claims-processing system that identified powered-wheelchair claims and tested them for possible abuse--before a claim was paid. That's unusual in the health-insurance industry, which is projected to pay out $1.8 trillion across 4 billion claims this year. In the huge consumer-credit-card market, sophisticated software looks for cheats before companies part with their money. But most public and private health insurers examine only a small number of claims after the fact, hemmed in by industry pressure to pay claims quickly and a lack of technological tools. By looking at claims before they're paid, Medicare has been able to stop payment on $140 million in suspect wheelchair claims. "We'd like to move more of our fraud-prevention effort into a proactive mode," Brandt says.
That wish may soon become reality for broader categories than motorized wheelchairs. Armed with new software tools, two insurers--Aetna Corp. and Vista Health Plan Inc.--are applying fraud-detection techniques, historically used in the credit- card market, to examining all claims before they're paid. Fair Isaac Corp., a maker of statistical credit-scoring software for credit-card companies, retailers, and insurance firms that reviews 65% of worldwide credit-card transactions for fraud, this month released a new version for the health-care industry of its Payment Optimizer software, which applies fraud-detection techniques used by companies such as Visa U.S.A. and MasterCard International Inc. to health-insurance claims. Vista, which insures more than 350,000 people in Florida, is testing Fair Isaac's system to review every claim before it's paid. Insurers use their claims-review systems--after payment--to look at only a fraction of the total, though that share can vary dramatically, from 5% to as much as half. "It's Fair Isaac's job to stay ahead of the criminal," says Bill Rushton, internal audit and fraud-prevention director.
Meanwhile, IBM is developing a software module for its Fraud and Abuse Management System, which is used by companies such as Horizon Blue Cross Blue Shield of New Jersey and Humana Inc. to review claims for possible insurance cheats. The module can review insurance claims for fraud before they're paid, and insurer Aetna plans to go live with it in October. When cheats are chased after they have their money, insurers are lucky to collect half of fraud losses. Instead, Aetna plans to unleash IBM's technology on up to 1 million daily claims that wend their way through Aetna's system, says business systems manager Ben Wright. "We think we will reduce the total loss," he says.
The software companies--and the insurers testing their latest technology--hope using prepayment analysis can reduce fraud levels to something closer to what they are in the credit-card industry: 0.06% of all transactions. Getting money back after it's been paid is more difficult--months of investigation typically lead to negotiated settlements or court dates that recover only 10 cents on the dollar. Sometimes, suspects simply walk away because enough evidence to indict can't be found.
Examining claims before they're paid "is the Holy Grail of the health-care industry," says Bill Mahon, who consults for insurance payers on fraud and is former president of the National Health Care Anti-Fraud Association, a nonprofit organization for sharing fraud information among insurance companies. Until recently, the industry took the point of view that examining claims before they were paid resulted in too little fraud detection and unacceptable delays. So for all its heavy-duty claims-processing systems and data-mining techniques, the health-care industry remains reactive to cases of fraud rather than trying to prevent them in the first place. Bilking Medicare, Medicaid, and private health-care plans "is a national epidemic," says Dave Hennings, 2004 chairman of the National Health Care Anti-Fraud Association. "We've been combating fraud from a reactive mode," he says. "IBM and Fair Isaac are moving to a more proactive mode, as in the credit-card industry."
The escalating costs of fraud may lead companies to look for new solutions. Estimates of the amount of U.S. health-care fraud range from 3% to 5% of filed claims. That translates into an estimated $57 billion to $94 billion in losses this year--equivalent to the annual tax revenue of the state of California at the high end of that range. And the problem is growing, at a 7% to 10% annual clip, according to Tim Delaney, chief of the Federal Bureau of Investigation's health-care fraud unit. By 2010, $154 billion in claims a year could be bogus. The problem is so persistent that some observers ask why the health-care industry can't do what the credit-card industry did over the last 12 years: cut fraud in half, or even greater, by using fraud-detection computer systems up front.
The Agile ArchiveWhen it comes to managing data, donít look at backup and archiving systems as burdens and cost centers. A well-designed archive can enhance data protection and restores, ease search and e-discovery efforts, and save money by intelligently moving data from expensive primary storage systems.
2014 Analytics, BI, and Information Management SurveyITís tried for years to simplify data analytics and business intelligence efforts. Have visual analysis tools and Hadoop and NoSQL databases helped? Respondents to our 2014 InformationWeek Analytics, Business Intelligence, and Information Management Survey have a mixed outlook.