Nationwide SIU
Contact | Investigative Services Order Form | Sitemap
Contact Us: 800-960-NSIU (6748)
Caselink

Overcoming Workers’ Comp fraud with detection technology

April 20, 2016 by  
Filed under Blog

Auto insurers get all the attention when it comes to fraud detection. It seems that no-fault states are the center of the fraud universe. But what about commercial claims?

The fraud problems faced by Workers’ Compensation insurers are significant, and they, too, can use fraud analytics programs to help detect bogus injury claims, fraudulent medical providers, and premium avoidance scams with outstanding results.

While there does not appear to be any commonly agreed upon measurements regarding the impact of Workers’ Compensation fraud, it is clear that the industry believes it to be a significant problem. A National Insurance Crime Bureau study indicates that Workers’ Compensation fraud costs insurers as much as $7.2 billion annually. The Coalition Against Insurance Fraud has stated that “Workers’ Comp fraud is a large crime in America today. Tens of billions of dollars in false claims and unpaid premiums are stolen every year.”And 69% of Workers’ Comp insurers expect a rise in scams.

Based on the financial implications and concerns about fraud being a growing problem, why do Workers’ Comp carriers appear to be underserved in the automated fraud detection area? One factor may be that they don’t perceive fraud and abuse to be a high priority compared to other issues like medical management, disability management and case reserving. When Rising Medical Solutions’ 2015 Workers’ Compensation Benchmarking Study asked participants to rank the core competencies most critical to claim outcomes, the Workers’ Comp carriers ranked fraud and abuse detection ninth out of the 10 categories listed.

It can be a balancing act to identify potentially fraudulent claims and still provide superior customer service for justified claims,…

Another explanation may be that Workers’ Comp carriers have just been slow to adopt automated fraud detection technology because of resistance to change, reliance on manual processes, limited IT resources, and concerns about data access and quality.

The same study also found that “As critical as technology is to maximizing claims effectiveness this is an opportunity area where the industry can make great progress. The 2013 and 2014 study results reflect that less than half of organizations are using workflow automation to manage best practices and about one quarter are using advanced analytics such as predictive modeling.” 

Clearly the Workers’ Comp world needs to change its mindset about, and approach to, detecting fraud. So what are the main drivers of Workers’ Comp fraud that insurers should focus on detecting?

Types of Workers’ Comp fraud

They tend to fall into three categories: employee claims fraud, medical provider fraud and premium avoidance.

Each of these categories have several associated scams. For example, in employee claims fraud, you will see fake injury claims, inflated injury claims, claims for injuries that occur outside of work, and malingering.

It can be a balancing act to identify potentially fraudulent claims and still provide superior customer service for justified claims,…

In medical provider fraud, you might see fake clinics with no licensed medical providers that are ground zero for bogus Workers’ Comp claims, or medical mills that enhance the nature and extent of injuries, overprescribe and overbill for treatments, or bill for treatments not rendered to line their own pockets and build up the employee’s workers’ comp claim.

Finally, there is premium avoidance fraud, which not only affects the insurer, but can also hurt the employee by costing them much needed coverage when a legitimate accident occurs. Examples of premium avoidance scams are misclassifying employees as having jobs that carry less risk (think office worker vs. roofer), and underreporting payroll and number of employees to gain lower premiums.

All of these scams are hard to identify when you factor in the massive amounts of data to be sifted through, the ever-increasing caseloads of adjusters and underwriters, state statutes that mandate time limits on certain claims functions, and the competing priorities of fraud detection and customer satisfaction.

The reality today for most Workers’ Comp insurers is that they are relying on manual detection processes and adjuster referrals to detect and feed suspicious claims to their special investigation units (SIUs) to investigate. It’s a well-known maxim in the SIU field that within claims organizations that rely heavily on manual adjuster-made referrals, 80% of your referrals come from 20% of the claims adjusters. This means there is a large number of missed fraud investigation opportunities that are hitting the insurer’s bottom line. So how can automated fraud detection capabilities help insurers overcome these challenges?

It begins with identifying your most valuable data to be used for fraud analytics. To effectively apply analytics, insurance companies need to integrate this data that resides in multiple internal-source systems as well as external third-party databases and pull it together for analysis.

To address the fraud scams discussed above, this would mean that Workers’ Comp carriers should (at a minimum) be looking at claims data, medical records, medical billing and bill audit data, pharmacy data, medical provider information, policy application data, underwriting data, loss history data, internal and industry watch list data of known bad providers, and any available “unstructured data” in the form of text such as claims or sales notes.

When the data is collected, the focus shifts to data preparation and cleansing, particularly entity resolution. When an insurer has an individual with profiles across multiple systems, it needs to be able to identify that as the same person and resolve data variations into a single entity to ensure accurate and effective analytics.

Once the data is collected and in a usable format, the insurer can begin to apply various analytic techniques to identify claims with strong indicators of fraud. Examples of analytic techniques used in fraud detection are heuristic rules, anomaly detection, predictive models, text mining and link analysis.

While all of these approaches will provide some degree of lift, the most comprehensive detection approach doesn’t rely on any one of these techniques, but instead combines them into a holistic approach. This will make the SIU less reliant on referrals from adjusters, agents or underwriters, and allow them to cast a wider net and capture all fraud investigation opportunities. This holistic approach ensures the acceptance of high-quality referrals that result in significant investigation outcomes, increased investigator efficiency, and a reduction in claims loss costs. It allows workers compensation insurers to assure their customers that they are aggressively combating fraud to help keep premiums competitive.

Comments

Tell us what you're thinking...
and oh, if you want a pic to show with your comment, go get a gravatar!





*