(Editor's Note: This article has been contributed by Janine Johnson, an analytics manager at the ISO Innovative Analytics (IIA) unit of Verisk Analytics (www.verisk.com).

It's no secret: Fraudulent claims continue to be an insidious problem for the industry, costing P&C insurers and consumers an estimated $40 billion a year (according to the FBI). Of course, that figure stands to skyrocket as the National Insurance Crime Bureau (NICB) reports that questionable claims (QCs) increased an unprecedented 20 percent during the first half of 2012 when compared to the same period in 2011.

Fraud is highly adaptive, responding to external stimuli and evolving over time. Historically, fraud has shifted from primarily vehicle thefts and property arson to the casualty side, as special investigative units (SIUs) clamped down on property scams. Casualty claims are notoriously more complex, involving collusion between physicians and lawyers and larger dollar amounts. These sums draw the attention of organized rings involving medical professionals and criminal syndicates.

As the battlefield continues to shift and grow, it is imperative for P&C insurance carriers to adapt faster than the fraudsters by employing cutting-edge tools and making fraud fighting integral to every point in the insurance transaction. The key elements of this strategy include:

Identifying potential problems during underwriting, preferably at time of application for policy.
Using claims triaging models to properly segment claims and reduce workload.
Adopting the most advanced tools and techniques to detect the truly suspicious claims.
Keeping an eye on future technologies, particularly those used in other disciplines.

Perimeter Defenses

The proliferation of direct-channel business creates opportunities for fraud that didn't exist prior to the viability of online policy processing. Early warning signs may include numerous applications coming from a single Internet protocol (IP) address. Additionally, IP addresses from unlikely places, such as states other than those in which a carrier does business, should raise a red flag. Insurers also need to evaluate the e-mail address supplied. E-mail addresses containing the applicant's name with corporate domains or well-known Internet service providers tend to be more credible than applications with a more anonymous and obscure points of origin.

Aside from those technological patterns, many traditional fraud indicators that are present at the time of a claim are also present at the time of application. Make use of this information to help identify policies that may be purchased simply to commit fraud.

No single indicator suffices to deny coverage, but applications that raise suspicion should be taken off of the direct-channel conveyor belt and placed in front of an underwriter, who will ask additional questions to verify information prior to granting coverage. In many instances, a dishonest applicant will not continue with the submission process when required to be in direct contact with an underwriter.

Claims Segmentation

Another proven fraud-fighting technique is to start by segmenting the legitimate claims. Doing that serves two purposes: First, the sooner a legitimate claim gets paid, the greater a customer's satisfaction with the company. Second, quickly identifying claims to fast track reduces the total number of claims that require adjuster or SIU scrutiny. This allows limited claim and investigative resources to focus on claims with some degree of suspicion—saving time and money by putting resources where they are needed most.

Step one in the segmentation process is identifying "contra-indicators" in the claim file. Contra-indicators are claim characteristics that indicate a claim is likely legitimate and may be paid as expeditiously as possible. For example, a police report, confirmation of transportation by an ambulance, or tests and images verifying an injury may all indicate a claim's legitimacy.

Cutting-Edge Detection Technology

Once legitimate claims are paid, remaining claims call for efficient fraud detection tools. Fortunately over the past decade, the reduced cost of data storage and the exponential increase in computing power have yielded the ability to leverage more information than ever.

Text mining has become more accessible because of increased computing. Among the most common applications of this technology is the mining of claims adjusters' diaries. Ideally, an adjuster should recognize and escalate the same suspicious indicators that text mining uncovers. Unfortunately, lack of training or time constraints can prevent an adjuster from picking up on those red flags. Text mining applies a consistent level of scrutiny to every diary entry, ensuring that even subtle indicators will not go undetected.

A second powerful technique that is rarely used is anomaly, or outlier, detection. The algorithms involved are computationally intensive yet extremely powerful for identifying behavior that does not fit the norm. Adopted from artificial intelligence, the algorithms learn new patterns to detect unusual characteristics even if the anomalous behavior does not include hallmark SIU red flags. This allows potentially problematic patterns to be identified earlier and put into the SIU pipeline for investigation.

The last decade also has witnessed a growth in the identification and analysis of networks, which has resulted in a broad area of research known as social network analysis. Harnessing that research for use in the P&C insurance industry provides the ability to identify fraud rings running various scams, including the following:

Staged accidents
Medical provider collusion
Attorney collusion

Network technology allows insurers to construct holistic claim networks proactively without the need for a known suspect starting point. Once constructed, the networks are scored using advanced predictive analytics that identify those that have unusual or suspicious characteristics. This technology will highlight networks before they fully metastasize by tracking various network velocity measures, such as changes in claim volume. This allows carriers to intervene much earlier in the process and avoid millions of dollars of fraudulent payouts.

Next Generation

Finally, claims executives need to keep abreast of tools and techniques used in other industries or disciplines. Often, best practices developed by others can be easily adapted quickly and effectively to the P&C industry.

Next-generation detection systems will help predict geographic migration patterns of fraud activity and identify false medical clinics attempting to mask their identity. They will include more and wider-ranging data sources, such as analysis of vehicle and medical images, automated use of social media, voice analysis, and GPS and cell phone data. To make the claims team as efficient as possible, always consider adopting the latest available technology from multiple disciplines.

The fight against fraud is a constant struggle to thwart ever-changing crooks. In order for carriers to adapt faster than criminals, they must adopt a multipronged approach to fighting fraud that relies upon vigilant detection at every stage of the insurance life cycle.

Special Reports /

Technology, Claims Segmentation Enhance Fraud Detection

Related Stories

Recommended For You