Printer Friendly

Fraud Detection: Taking a Hard Stance in a Soft Economy.

The good news is: We're making progress in the fight against insurance fraud. The bad news is: Due to the softening economy and other factors, incidences of fraud are expected to increase. Fortunately, new applications of intelligent decision management software employing advanced artificial intelligence technology are empowering the investigative teams with the ammunition they need to take the fight on fraud to the next level.

The efforts by insurance carriers and federal and state governments to find and prosecute insurance fraud are starting to produce results in recovering losses and in deterring potential fraud perpetrators. However, a number of factors in the environment and the marketplace are changing, which could create an upswing in insurance fraud in the near future.

As the economy softens, opportunistic fraud is expected to increase. Workers facing layoffs may be tempted to stage automobile accidents, file false claims, or extend their disability periods. Providers and/or attorneys with declining revenues may resort to upcoding, providing unnecessary services, or extending the life of a case. Corporate managers, looking to improve their bottom lines, may "fudge" on the classification of workers or over-report damages for claims.

The use of the Internet to process transactions in real-time could exponentially increase the opportunities and scope for fraud perpetration. When a transaction involving thousands or millions of dollars occurs in a second, the theft of those amounts can occur just as rapidly. Ever-resourceful criminals --including organized crime -- are apparently turning their brainpower toward new schemes to illegally profit from health care and insurance scams. Even seemingly benign trends, such as the growth of no-fault auto insurance regulations, can create new opportunities for crooks. Since New York adopted its no-fault auto insurance program, fraudulent auto claims have increased by as much as one-third.

Stronger laws and rigorous investigation and prosecution are acting as a deterrent, and the industry's efforts to educate consumers are important -- especially since most fraud detection today is accomplished by alert claims adjusters and whistleblowers. But it is time to move to a new level of attack.

As noted in the recent fraud survey by Conning Corporation, the next level of fraud detection must rely on technology to automate and systematize the process. The survey estimates that only about 20 percent of fraud is detected; probably significantly less is prosecuted and covered. Out of 116 million claims reported in 1999, there were only 10,000 SITU investigations, representing .000086 percent of all claims -- nowhere near the estimated 10 percent of claims that are fraudulent.

Why isn't more fraud detected, and how can technology remedy this situation?

The first barrier to improving detection lies with the claims adjusters. Adjusters, typically, do not have incentives to fight fraud. Fraud detection is usually secondary to confirming coverage and making payment deadlines. Busy claims adjusters are often rewarded based on productivity in processing and closing claims. They simply don't have time to sift through claims data to find patterns that are suspicious. Besides, the vast mountain of claims data is too large for any human brain to process.

Second, the insurance claims process is still primarily paper-based, allowing for significant informational "disconnects" between the parties, and for information to slip through the cracks. This situation prevents anyone from seeing the entire picture of a claim at any one time.

Third, the effectiveness of special investigators is directly proportional to the quality of the leads they receive. With today's available information, a claim may look suspicious but be perfectly legitimate -- yet further investigation is required to determine its validity. Adjuster and investigator time is wasted in false leads. And when a fraudulent claim is detected, it is usually too late in the process to stem most of the loss.

Fourth, fraud perpetrators are like Internet "hackers" -- as fast as one hole of opportunity is plugged, a new scheme is hatched for which there is yet no remedy.

Consistency, accuracy and speed can now be incorporated into the fraud detection process.

Through the use of new technology -- which is powered by the artificial intelligence in neural network models and embedded in new, intelligent decision management software -- adjusters can be notified immediately and continuously about suspicious claims that are constantly evaluated based on the patterns created by thousands of pieces of information. Routine medical bills that result from a claim can be automatically paid without requiring any human review. Of the remainder, those that might be fraudulent can be automatically routed for special handling -- again without human intervention required to make this analysis.

This will occur when a claim is initially filed, or further on in the life of the claim if it appears that opportunistic fraud has reared its head. Thus, fraudulent activity can be caught at the earliest possible moment in the claim's life cycle. It can be found consistently, and not be dependent on a super-alert adjuster or a man-on-the street whistleblower.

The result is that fraud can be detected early in its life cycle and qualified leads can be generated for investigators, while manpower required for the detection process can actually be reduced.

These new systems can work with a carrier's existing legacy system, eliminating the costly expense of replacing proprietary and existing software, and capture the collective intelligence throughout the claims department by the combination of rules and artificial intelligence in the system. The novice adjuster has access to the same body of knowledge as the veteran, thus adding consistency and compression of the cycle time to fraud identification. These smart systems will find new schemes of fraud and abuse before millions are lost -- very important in today's environment when professional criminals see insurance fraud as a gold mine of opportunity and are creating new schemes as old ones are foiled.

With these applications, carriers can structure their computer systems so workflow and communications are available in a more efficient database that can be accessed in real time by claims and SIU personnel. Finally, advanced fraud detection software can be predictive and able to prospectively identify fraud - an essential characteristic for electronic fraud detection.

The fraud detection technology, coupled with the new intelligent decision management software, delivers a one/two punch to fraud perpetrators.

Will technology eliminate the human component in the war to stop fraud and abuse? Of course not. Criminals must still be caught and prosecuted by people, not computers. Insurance companies and third-party' claims administrators must continue to properly staff and train the claims and SIU operations, and states must support active prosecution of fraud. Professional associations should continue their efforts to raise public awareness and sway public opinion. What this new technology will do is find existing fraud faster and more accurately that other methods, taking the fight against fraud to the next level with the "good gays" having the advantage.

John D'Alusio is senior vice president of HNC Software, San Diego, CA. The company's sophisticated software is used in numerous industries, including telecommunications, financial services, insurance, and e-commerce.
COPYRIGHT 2001 ALM Media, LLC
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2001 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:D'Alusio, John
Publication:Claims
Date:Oct 1, 2001
Words:1153
Previous Article:Unexamined Business Activity Isn't Worth Doing.
Next Article:Are Insurance Adjusters Professional?

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |