The quiet catastrophe of insurance fraud is gaining more attention as insurance executives continue to look to operational efficiencies-rather than investment income-to protect their bottom lines.With the Coalition Against Insurance Fraud (CAIF) estimating an annual fraud cost of $80 billion dollars, the industry has realized that the harmless fudging of a million here and a million there is adding up to real money.
The temptation of the Internet-and the new fraud opportunities it presents-adds to this worsening problem. Underscoring the nature of the battle to fight fraud, a recent Accenture survey gauged that nearly one in four U.S. adults believe that overstating the value of claims is acceptable.
Unfortunately, the most common method to date for fighting fraud has been discovering fraud by accident, with few proactive steps having much impact.
Given the attitude of the consumer and the clever methods of fraud perpetrators, getting the upper hand in this game of one-upmanship will require a combination of commitment to the fight, continuous process improvement and effective technology implementation.
Fighting a losing battle
It's not accurate to say that the industry has been sitting back and letting the fraud wizards run rampant through the industry. Nevertheless, the early technology approaches had the industry in a position of taking one step forward and two steps back, continuing to lose ground despite valiant efforts.
Early technologies allowed for rudimentary hard-coded programs to detect known suspect circumstances. However, it didn't take long for fraud perpetrators to catch on when a certain medical claim of more than $100 would be flagged as suspicious. Their response was to submit multiple claims for $75 and merrily continue on their fraudulent journey.
The next-generation of fraud detection tools during the 1980s and 1990s brought more hope with relational databases and analytical tools, such as query languages to perform ad hoc 'what if' and 'if, then, else' tests against a high volume of data. These rules-based approaches proved useful, but analysis and detection were still limited to looking for the 'known problem' or those the business analyst could conceive of and search for.
When fraud patterns were detected, it would often take a year or more to make the necessary program and process changes. By then, the criminal had already moved to fresh and fertile ground of new fraud opportunity.
The biggest stumbling block of the earlier technologies was that they tended err on the side of caution, flagging any suspicious claim, which would then require a subsequent manual investigation. Time spent investigating flagged claims that later proved to be valid resulted in unproductive use of time, and raised the ire of legitimate claimants.
The technology tool sets and software available today are far more advanced in computational techniques and have the ability to 'learn as they go,' enabling them to adapt to the ever-changing fraud environment, and increasing their sophistication over time. The more 'intelligent' nature of these tools reduces the reliance on the human element that has been one of the limiting factors with early generation fraud-fighting methods.
By attacking fraud through the use of advanced predictive analytics and adaptive optimization techniques (rough sets, classifier systems, revolutionary programming) combined with early generation neural networks and rules-based tools, the industry will experience a significant increase in fraud detection accuracy and timeliness-and a decrease in the occurrences of false positives.
In addition, the self-educating nature of the tools will improve the results and offer the ability to identify new fraud trends virtually as they are happening-allowing the insurance industry to finally get ahead of the bad guys.
However, even with the best technology tools, information in the form of data is still key to fighting fraud-and it is no secret that the insurance industry has less-than-perfect data tucked away in a myriad of silo systems throughout the enterprise.
Although technology can help organize the data and provide a common view, it is the commitment to improvement that will ultimately lead to the data completeness, accuracy and quality that it is the ultimate objective.
More importantly, carriers must start now and improve the data over time, rather than waiting for perfect data before beginning to fight fraud.
While the feet on the street for the investigative claims work will still be critical to success, a reliance on today's more sophisticated technologies will provide fraud investigators more accurate information, making their time spent in investigation more productive than chasing down a high volume of false positives.
There is no better time than the present to wage the war on fraud. Based on the Accenture survey results, it appears that consumers will continue to take a passive approach.
And, while carriers continue to wait for state legislatures and Congress to enact tougher fraud laws and establish more fraud bureaus, insurers-and ultimately policyholders, in the form of higher premiums-continue to pay the price.
Eric Miller, CPA, CFP, is senior principal with Highpoint Partners LLC, a Charlotte, N.C.-based management and technology consulting firm.
Register or login for access to this item and much more
All Digital Insurance content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access