​Targeting Health-care Fraud

U.S. investigators are using advanced analytics to detect fraudulent health-care practices.

Comments Views

​The U.S. Justice Department (DOJ) has charged 601 people with health-care fraud, as part of the department's annual takedown of fraudulent health-care activities, Reuters reports. Officials estimate the frauds amounted to more than $2 billion in losses. This year's takedown announcement focused on the opioid crisis, with more than 160 doctors and other suspects charged with prescribing and distributing addictive painkillers.

Lessons Learned

This story represents both good and bad news. The good: An impressive series of successful fraud investigations — the largest single medical fraud enforcement action in DOJ history — resulted in hundreds of charges across the U.S. The bad: The announcement illustrates how health-care fraud remains a persistent and costly problem for governments, the private sector, and taxpayers.

To address health-care fraud, the U.S. Congress and the Center for Medicare & Medicaid Services (CMS) have developed a variety of approaches in recent years to audit Medicare and Medicaid claims and detect fraud. Two of these programs, the Fraud Prevention System (FPS) and Comprehensive Error Rate Testing (CERT), appear to have been instrumental in the government's latest efforts to uncover fraudulent activities. Here is a closer look at how they work.

FPS. CMS' advanced analytics system, FPS, uses predictive analytics to identify troublesome billing patterns and outlier claims for action, similar to systems used by credit card companies. Predictive analytics is a branch of advanced statistics that uses historical data to make predictions about future events. FPS uses predictive analytics based on detection methods such as coding rules, anomaly detection, and link analytics involving specific algorithms (based on regression routines, nearest neighbor and neural networks, and similar algorithms) to associate scores to likely matches that indicate fraud issues.

Reviews may be fully automated, such as analyzing 100 percent of Medicare fee-for-service claims. Alternatively, they may be semi-automated. For example, for high-scoring claims, FPS may link the National Provider Identifier to Tax Identification Number codes to identify a specific entity and associated billing and taxation activity, and detect anomalies.

CERT. This program randomly selects a sample of claims submitted to insurance carriers and Medicare Administrative Contractors (MACs) during each reporting period. CERT then requests medical records from the health-care providers that submitted those claims. By reviewing claims in the sample and the associated medical records, CERT can see whether these claims complied with Medicare coverage, coding, and billing rules. If they did not, it assigns errors to the claims.

Although there are a substantial number of types of claims, globally error rates are in the 10 percent to 12 percent range, amounting to billions of dollars each year. This does not necessarily mean there is fraud involved. Where the provider did not submit medical records, CERT classifies the case as a no-documentation claim and counts it as an error. It then sends providers overpayment letters or makes adjustments for claims that were overpaid or underpaid. In some cases, CERT may notify providers that further investigation — including for fraud — may be underway. Some of the key red flags include:

  • An invoice with a modified date.
  • An entire family claiming similar supplies or services.
  • A history of frequent or high-value claims.
  • Many plan members in one group using the same health-care provider or trends.
  • A plan member consulting many health-care providers or buying drugs at numerous pharmacies.
  • Vague or evasive answers to questions by plan members or health-care providers.

Beyond the FPS and CERT programs, the medical and pharmaceutical industries, associations, professionals, and patients, as well as potential fraudsters should take note of the DOJ's recent enforcement action. They should continue to improve their fraud awareness and compliance behavior — or risk getting caught.

Art Stewart
Internal Auditor is pleased to provide you an opportunity to share your thoughts about the articles posted on this site. Some comments may be reprinted elsewhere, online or offline. We encourage lively, open discussion and only ask that you refrain from personal comments and remarks that are off topic. Internal Auditor reserves the right to remove comments.

About the Author



Art StewartArt Stewart<p>​Art Stewart is an independent management consultant with more than 35 years of experience in internal audit, financial management, performance measurement, governance, and strategic policy planning.​​​</p>https://iaonline.theiia.org/authors/Pages/Art-Stewart.aspx


Comment on this article

comments powered by Disqus
  • Galvanize-September-2020-Premium-1
  • FSE-September-2020-Premium-2
  • Auditboard-September-2020-Premium-3