Technology increasingly driving insurer anti-fraud strategies

By Dennis Jay & Stuart Rose
September 11, 2012

Analytics help combat fraud rings, but many insurers lack advanced tools

JIFA_logoAbstract: Anti-fraud technology has been a major contributor to the maturation of insurer anti- fraud programs in recent years. The Coalition Against Insurance Fraud undertook a study to better understand insurer adoption and use of technology. The technology company SAS assisted in the survey of 74 insurers. Nearly all insurers use technology, but vary widely in the sophistication of their systems. Less than half use advanced technologies such as predictive modeling and data mining. The biggest challenges to technology deployment are the need to demonstrate a clear return on investment, and lack of IT resources. Technology isn’t likely to replace the professionalism, instincts and street experience of investigators, but can greatly improve the investigative process. Insurers seem to recognize the importance of technology; less than five percent of respondents expect a decrease in technology funding. Going forward, technology especially will be needed to combat large organized rings, which are a growing insurer focus. The good news is that insurers are planning to invest more in technology.

Insurer anti-fraud programs have matured greatly over the last several years. They are becoming more sophisticated, efficient and effective in detecting and investigating insurance fraud. Anti-fraud technology is a major reason for this success and increasingly is seen as integral in the battle against insurance crime, especially organized rings. But major challenges remain for insurers to fully deploy technology solutions, according to a new study[1] conducted recently by the Coalition Against Insurance Fraud with the assistance of the technology company SAS.

The study sought to take a snapshot of the degree to which insurers use technology, the types of analytic systems they use, the obstacles insurers face in using technology and the benefits they gained. The study consisted of an online survey in which 74 insurers participated, interviews with individual insurers and a review of existing literature.

Nearly 90 percent of insurers surveyed are using anti-fraud technology, yet most still are only using basic tools such as automated red flags, claims scoring and link analysis. Less than half of insurers surveyed are employing predictive modeling, text mining, geographic data mapping and other advanced analytics.

Only about 14 percent said they use any automated tools to detect underwriting or point-of-sale fraud. Anti-fraud technology is maturing rapidly and many insurers have come a long way in the last few years. But the study suggests many insurers have much room to grow.

SIUs drive tech expansion
JIFAstory4While the vast majority of survey respondents (79 percent) say that their Special Investigation Unit (SIU) is the primary sponsor of anti-fraud technology initiatives, the departments often funding these projects were other internal units such as claims, IT and corporate enterprise. This suggests that while the SIU has responsibility for addressing fraud risk, it may not control the budget to implement its own technology projects. While this close collaboration between departments is expected, in some organizations it may hamper the speed with which technology can be deployed due to additional approval processes.

Most respondents agree that technology-based solutions can provide compelling advantages in fraud detection, but insurers still face many challenges with technology projects. In fact, 74 percent of respondents say the biggest challenge is either “Lack of IT resources” or “compelling cost/benefit analysis.” Despite these challenges, it seems that technology is being adopted at an increasing rate. More than half of the insurers surveyed say they have begun using anti-fraud technology solutions within the last five years, many within the last two years.

More insurance companies are relying on solutions from vendors instead of internal IT teams. Half of survey respondents use solutions built by a vendor, and one-quarter use a solution hosted by the software vendor or third party. Common reasons for this include limited capacity of internal IT resources, and an interest in reducing the timeline to adopt, learn and start using the software. Notably, several respondents cited legal and compliance challenges to getting a fraud technology solution deployed. This suggests that some insurers may still be struggling with these hurdles [see Figure 1].

When it comes to technology, most insurers’ anti-fraud efforts begin with rules-based systems. These systems test and score each claim against a predefined set of business rules and report the results to the SIU teams that look suspicious due to their aggregate scores or relation to threshold value. The advantage of the business-rules approach is its simplicity. Unfortunately these business rules often generate high false positive rates and quickly become obsolete because swindlers can easily learn and manipulate rules to their advantage.

Technology is unlikely to eliminate claims adjuster intuition and SIU street expertise, instincts and professional judgment. But it certainly can help improve the investigation process. Fewer false positives, more-accurate fraud detection, fewer stolen insurance dollars and enhanced customer satisfaction are a few of the valuable returns on investment in anti-fraud technology. Effective systems also can give insurers confidence in paying legitimate claims more promptly, thus shortening claim cycle time and reducing the time legitimate claimants have to wait to be paid.

Current tools & technology
Common wisdom in the industry says any effective anti-fraud program should include at least the basic technology to detect and investigate fraud. Yet, 12 percent of respondents indicated they deploy no anti-fraud technology.

For organizations that embrace technology, nearly all say they use it for claims fraud detection and investigation. Most insurers use a combination of anti-fraud technologies.

The top three technologies that survey respondents use are automated red flags/business rules (64 percent), scoring capability (60 percent) and link analysis (57 percent). In most cases these tools automate many of the manual tasks (i.e. business rules) associated with fraud detection. But less than half of insurance companies use more advanced techniques such as workflow routing (43 percent), text mining (40 percent), predictive modeling (40 percent) and geographic data mapping (23 percent).

While technology has come a long way as an anti-fraud tool, it still relies heavily on the volume and quality of the data used in the analysis. Unfortunately, information silos are still prevalent in the insurance industry. Many organizations use a combination of multiple legacy systems, spreadsheets and external databases.

Unsurprisingly, the most common source of information by far was the carrier’s own claims data. Other popular data sources used by the majority of survey respondents were industry claims history data (69 percent) and public records (62 percent). Industry fraud alerts or watch list data are used by 57 percent of responding insurers [see Figure 2].

Studies suggest at least 75 percent of the information available in an organization is unstructured data, and that percentage is likely to increase with the growth of social media[2].

Only 40 percent of respondents say they use text mining as an anti-fraud tool. Additionally, 36 percent report that they investigate social-media data for case leads. This figure is likely to increase as more insurance companies embrace this growing trend.

Benefits and future investment
“Effective use of technology gives many insurers more confidence in their anti-fraud programs. Insurers increasingly understand that investing in technology can earn a positive financial return.” The current economic climate continues to pressure many organizations to reduce expenses. Fortunately, insurance organizations seem to recognize the importance of fraud detection and its impact on the bottom line. Less than five percent of respondents expect a decrease in their anti-fraud technology budget during the next 12 months, and nearly one-third expect more funds.

Where are insurers spending this funding? The answer includes different technologies, but the primary investment over the next 12 to 24 months includes predictive modeling (33 percent), text mining (31 percent) and automated red flags (25 percent).

The historically high frequency of false positives from manual red flags and automated business rules has created a lack of confidence in “flagged[3]” occurrences, and challenged insurers to balance between investigating suspicious claims and ensuring convenience for legitimate claimants.

One answer may lie in getting smarter about fraud detection and leveraging more advanced technology. In fact, 57 percent of responding organizations say higher quality referrals are the primary benefits of anti-fraud technology.

Several SIU directors say their current systems generate better-quality referrals and discover them earlier in the claims process, resulting in a better return on investment for their SIU team. In this case, technology helps enable a more proactive approach to fraud detection, resulting in improved impact rates and averted payment of fraudulent claims [see Figure 3].

A central theme is emerging in the insurance fraud space. While individuals continue to commit fraud reactively as opportunities arise, the risk from organized fraud activity continues to loom as a mounting problem. There is growing evidence that organized criminal enterprises, both loosely knit and sophisticated, continue to commit fraud with growing efficiency. This is true for no-fault auto and health coverage, and workers comp premium and claims fraud.

The size and scope of these operations were underscored earlier this year when one no-fault clinic ring in New York allegedly filed nearly $400 million in false claims between 2007 and 2012. This study confirmed the growing insurer focus on organized fraud. Half the respondents say that medical provider fraud, and 38 percent answered that PIP/no-fault fraud has the greatest impact on their companies.

Today, insurers, law enforcement and anti-fraud groups such as NICB are focusing more resources on medical-provider fraud, staged crashes and other organized fraud activity. Special investigation units within insurance companies are following suit in technology, in forming major case units and in providing enhanced training.

A major challenge in detecting organized fraud is to ensure claims are not viewed in isolation. With multiple business units using different systems, some insurers often struggle with connecting the dots to identify these fraud rings. To combat this growing problem, more insurance companies use link analysis and social-network analysis tools. In fact 52 percent of survey respondents say the major benefit of anti-fraud technology lies in uncovering complex or organized fraud activity.

More confidence in technology
Effective use of technology gives many insurers more confidence in their anti-fraud programs. Insurers increasingly understand that investing in technology can earn a positive financial return. In the process, these insurers are gaining more confidence that SIU leaders are key contributors in protecting corporate assets and policyholder value.

In recent years, insurers also have become more comfortable with technology overall as more aspects of their operations are driven by technology solutions. But such comfort wasn’t always the case in the anti-fraud arena. Early adopters suffered through hit-or-miss systems that produced far too many false promises and didn’t perform as advertised. Capabilities often were more over-promised and under-delivered, and few systems integrated well with existing technologies.

For most insurers, those days are long gone. Automated systems now are integral to their operations and vital to success as much as traditional investigation techniques. This study underscores the wide use of technologies for detecting claims fraud and conducting investigations. A majority of insurers have at least five years experience with such systems, and interviews suggest an overall satisfaction with their technology. But a sizable segment of the industry has yet to embrace anti-fraud technology either at all, or fully. Many are still running antiquated legacy systems or haven’t integrated anti-fraud technology with claims and other systems. And most insurers still haven’t adopted technology to combat premium fraud and other non-claims swindles.

The good news is that many insurers say they plan to invest further in technology in the future. Hopefully that trend will spread to other late adopters. With an expanding market, technology companies have more incentive to innovate and introduce more product lines. Additionally, the substantial investment of the federal government in technology to detect and prevent healthcare fraud likely will spur innovation and perhaps even help lower the cost of future systems.

About the authors: Dennis Jay is executive director of the Coalition Against Insurance Fraud. Stuart Rose is global insurance marketing manager for SAS, a leader in providing business analytic software and services.

[1] State of Insurer Anti-fraud Technology, Coalition Against Insurance Fraud, September 2012
[2] Extracting Value from Chaos, IDC iView, June 2011
[3] Flagging Fraud – Insurance Networking News, August 2012

HTML Comment Box is loading comments...


Use your browser's back button to return to your previous search.

Search our articles database



Word search: