Verisk Health's Fraud Alliance: The Power of Pooled Data

By Dr. Karthik Balakrishnan

A family practitioner was billing a national healthcare plan, on average, for five hours of services rendered each day. Because five hours of treatment is reasonable and possible, the billing raised no red flags and was continually approved for payment. What the health plan did not know, however, was that the same provider was billing five other plans — a combined 50 hours of service on average per day. In reality, that would be impossible to perform without additional practitioners or physician's assistants, which he did not have.

That case is just one of many scams that plague the healthcare system today. In an industry constantly facing new schemes and emerging trends, preventing healthcare fraud, waste, and abuse (FWA) demands intense vigilance and evolving technology. While significant progress has been made in the fight against FWA, many schemes such as the one above simply cannot be detected in a single payer's data. The next step in fighting healthcare fraud will be to pool and aggregate data from multiple payers to create comprehensive views of providers and develop sophisticated analytics to discover suspicious and abnormal patterns in cross-payer data. That is precisely what Verisk Health intends to achieve with its latest effort: an alliance aimed at pooling healthcare data across multiple payers.

In developing its approach, Verisk Health's Payment Accuracy team drew inspiration from its corporate DNA, specifically in the property/casualty business. For decades, Verisk Analytics, through its ISO subsidiary, has specialized in building industrywide databases and overlaying sophisticated analytics for risk assessment and decision support. The company receives claims from more than 90 percent of the property/casualty industry, which it uses to detect suspect entities and fraud rings. Recently, Verisk Health joined ISO and the National Insurance Crime Bureau (NICB) to launch the Aggregated Medical Database (AMD) for the identification of cross-carrier medical fraud and aberrance patterns — an initiative that Risk & Insurance magazine recognized with its Risk Innovator™ Award.

Why an Alliance?

According to the National Health Care Anti-Fraud Association, health insurance fraud, waste, and abuse cost the country's healthcare system as much as $300 billion every year. Despite the growing industry demand for more robust detection and prevention efforts, the U.S. Department of Health and Human Services recently announced it may be facing staff and budget cuts, which opens a wide door for those actively seeking opportunities to defraud our healthcare system, particularly government-sponsored programs such as Medicare and Medicaid.

Furthermore, post-pay recovery — the act of recovering fraudulent or excessive healthcare dollars after a claim has been paid — represents a continuing cost to the industry and compounds a health plan's loss substantially.

It's impractical and inefficient for the Centers for Medicare and Medicaid Services (CMS) or private health plans to discover suspect entities on their own and then chase paid dollars. Instead, Verisk Health's alliance of healthcare payers is taking a revolutionary step by pooling data to uncover such schemes before claims are even paid. (See Figure 1.) Additionally, pooled data enables a 360-degree view of providers, allowing a unique and unparalleled understanding of billing practices, known schemes, and emerging trends. With those advantages, health insurance plans and their special investigative units are able to develop ­comprehensive detection and prevention plans and minimize their exposure to unnecessary spending.

Figure 1
Current and Projected Provider Overlap

The map below shows the percentage of current overlap in providers based on Verisk Health alliance membership. The second map suggests the increased overlap with the addition of another health plan currently in discussions to join.

The Alliance Emerges

Joining Verisk Health in the fight against FWA, Humana and Centene are among the first healthcare plans to pool their data and begin sharing best practices and suspect providers. "It's time to recognize that no one entity can be nearly as successful alone as we can be together," said Robert Miromonti, vice president, Ethics and Compliance, at Centene Corporation. "This is definitely the case with fraud, waste, and abuse, and that's why we're excited to be members of this innovative initiative. We're confident that a cross-payer, data-driven solution will prove invaluable in the fight against fraud."

Data Changes the Game

The notion of pooling data has historically been dismissed as unfeasible because of the challenges and restrictions in sharing healthcare information. Claims and submission data is highly protected. In addition to HIPAA mandates, numerous state insurance and privacy laws govern permissible use of the data and impose restrictions on information sharing and storing protocols.

"To date, the concept of pooling data across payers was considered too time-consuming and difficult to navigate because of the restrictive privacy laws in healthcare," explains Verisk Health CEO Joel Portice. "Because of this, we've spent countless hours assessing the risks of working with cross-payer pooled data and have gathered opinions from multiple HIPAA attorneys and statutory and regulatory law experts to be absolutely certain we're fully compliant with existing government mandates."

Designed and executed carefully with those considerations in mind, the resulting alliance is one giant step forward in redefining fraud detection and prevention. By fusing its sophisticated fraud detection software with antifraud research and analytics models and by employing investigative best practices in the alliance, Verisk Health has delivered the healthcare industry's first true cross-payer collaborative model.

Expanding Efforts

Verisk Health built the pooled-data alliance as an extension of its existing FWA solution, Fraud Finder Pro®. However, to accommodate the data and scalability needs of a true aggregated cross-payer solution, the team had to build the solution from the ground up using big data technologies such as Hadoop, Hive, and PMML (Predictive Modeling Markup Language).

In addition to technology and breakthrough analytics and models, the alliance will include a series of tools and capabilities such as case tracking, link analysis, and cross-payer communication tools to increase efficiency and joint efforts among its various associates. Through ongoing participation and member round­table sessions, Verisk Health hopes to improve the collaboration that is vital to a successful antifraud effort.

By fusing data, analytics, technology, and expertise, Verisk Health and its associates are unlocking data from the silos of individual payers — and eventually across multiple verticals — to detect fraud, waste, and abuse schemes as they emerge and prevent known schemes from ever affecting the bottom line.

Karthik Balakrishnan, Ph.D., is senior vice president of fraud solutions and analytics at Verisk Health.