Preventing Healthcare Fraud with Big Data and Analytics

Written by Mark Isbitts, Director, Market Planning, Payment Protection, LexisNexis Risk Solutions
  1. Home
  2. Insights and Resources
  3. All Articles
  4. Preventing Healthcare Fraud with Big Data and Analytics

fingerprint scanHealthcare fraud is a national problem, prevalent in federal and state, as well as private insurance programs. Over the last decade, healthcare fraud has skyrocketed with billions of dollars being paid on improper claims. The National Healthcare Anti-fraud Association conservatively estimates that three percent of all healthcare spending, or $60 billion, is lost to healthcare fraud. Other estimates place this number closer to $200 billion.

The increased incidence of identity theft is another major problem – more than 1.5 million people have been victimized by medical identity theft at an average cost of $20,000 to the victim. These statistics represent avoidable healthcare costs that directly impact the cost and quality of healthcare for every American. Healthcare fraud, waste and abuse (FWA) not only contributes to higher insurance premiums; every dollar spent on fraudulent or abusive claims reduces the amount of money available to improve the quality of care for those incurring legitimate expenses.

Now more than ever, the healthcare industry must migrate to a fraud control model that integrates fraud prevention and detection at the front-end of the payer workflow, applies analytic controls throughout the workflow, and incorporates post-pay detection and recovery processes at the back-end of the work flow. Claims review processes that incorporate rules-based data analytics, predictive modeling, and linking technologies allow commercial and government payers to identify fraud before an ineligible claim is paid. Effective fraud detection is best achieved through a layered approach to claims analysis, including identity analytics, claims analytics (predictive modeling and rules-based fraud detection), and social network analytics.

  • Identity analytics: A “true” identity management approach includes verification and authentication of all provider identities, and a commitment to ongoing background evaluations. A health plan’s current provider file can be quickly and easily evaluated for derogatory information or indications of risk. New providers should also have their identities verified and evaluated as part of a standard enrollment qualification process. All providers’ identities should be periodically reviewed for critical changes between enrollment periods. This will enable the recovery of mispayments and eliminate unqualified and risky providers from a payer’s network proactively.
  • Claims analytics: Comprehensive pre-pay claims analytics consist of rules-based screens and edits, combined with predictive modeling that identifies the potential for improper payments by “scoring” claims and/or providers before a claim is paid. Traditional rules-based fraud detection systems that analyze claims and identify outliers are most frequently deployed post-payment. Moving this operation to the front-end of the claims payment process and complementing it with predictive modeling techniques is key to changing the game in favor of providers that are giving proper care and payers who are attempting to pay legitimate claims as quickly and efficiently as possible.
  • Social network analytics: Much of the FWA that plague healthcare payers is the result of organized, sophisticated and collusive activities among providers and between providers and patients. Social network analysis can help identify relationships, links and hidden patterns of information sharing and interactions within potentially fraudulent clusters, including:
    • Patient relationships with known perpetrators of healthcare fraud;
    • Links between recipients, businesses, assets and relatives and associates;
    • Links between licensed and non-licensed and sanctioned providers; and
    • Inappropriate relationships between patients, providers, employees, suppliers and partners

As schemes and various provider technologies become more complex, healthcare organizations need to create wide-reaching FWA prevention programs that address the overall problem more holistically. By combining identity and entity resolution, rules-based claim and clinical review, complex linking analysis and predictive analytics into a seamless workflow, we will come closer to migrating an integrated pre-pay fraud solution to a real risk control environment with the potential to eliminate billions of dollars in improper payments due to FWA. This is not just a healthcare imperative, but a national economic imperative that must be addressed immediately. The analytics exist. It is time for those analytics to be implemented and the hard choices that enable that implementation to be made to insure that we remain at the forefront of quality care for all Americans.