Stopping Medicare Fraud - CMS Predictive Modeling Initiatives

When CMS releases news that a medical practice or DME provider has been busted for submitting millions of dollars in fraudulent claims, it makes many legitimate physicians and their staff wonder "How did that happen when we can't even get paid for a new patient visit or test?" The reason is that CMS is stepping up efforts to ferret out fraud through its new Center for Program Integrity (CPI), whose mission is to watch suspect providers/suppliers, their coding, and their claims.

Stopping large scale fraud is the primary objective of the CPI. Using technology developed by banks and credit card companies, CMS, in cooperation with Northrop Grumman, has put together a predictive analytic system using sophisticated algorithms that evaluate claim patterns by beneficiary, provider, and service origin. Those claims are then assigned "alerts" and "risk scores." Skilled analysts then review the claims. Their review may be shared with the MACs and Zone Program Integrity Contractors to "enact targeted payment denials." Innocent billing errors, on the other hand, will clear and payment will be sent.

Ted Doolittle, deputy director for the CPI makes it clear: "The central game is not about stopping a $500 claim. It's about getting a provider out of the program. If you get the provider out, their operation comes to a standstill." ¹

The CPI is after the "bad actors," not legitimate practices that may make billing errors in good faith. One interesting area is on the enrollment side where increased attention is being paid to the legitimacy of the provider on the front end.

Most practices are not intentionally sending out erroneous bills for services they did not perform, or committing other types of Medicare or Medicare fraud. Still, audits at other levels are taking place. While you are not likely to encounter the CPI auditors, the likelihood of hearing from a Recovery Area Contractor (RAC) is quite high. Seeking to recoup "overpayments," RAC auditors often focus on auditing E&M services - looking for overcoding.

So what can you do to protect your practice against negative audit findings? Review coding and chart documentation for all physicians in your practice. There are a variety of ways to accomplish this. You can have the practice administrator conduct random chart audits on a rolling basis, per physician; bring in a coding consultant to conduct a practice-wide chart audit; or examine your top E&M codes to look for trends such as consistently undercoding or overcoding. One tool you can use is the E&M Analyzer at www.karenzupko.com; you can profile a medical group and individual physicians against Medicare data for other physicians in your specialty and state. This comprehensive tool facilitates easy analysis of E&M data for possible "outliers."

In the chart below, one physician, who we will call Dr. Plas Tick, exceeds both national and state norms for new patient level-four visits, across his specialty. Should this physician be worried? No, not necessarily. But, he will benefit from a random chart audit to make sure his documentation for these visits support level-four codes. Often, use of a new EHR will result in a sudden jump in higher-level codes. It is important for practices to assess their EHR coding accuracy, especially if they use an "auto-coder."

In this chart, another physician, Dr. Ankles, falls in line with the national and state benchmarks for others in her specialty and is less likely to experience a RAC audit.

The desire to bill higher-level codes and increase practice revenue is understandable but is not worth the risk of incorrect coding and the audits that come with it. Code to the best of your abilities, run the E&M reports often, and be sure your claim submissions are as accurate and detailed as possible. This will minimize the risk of receiving a high-risk score and receiving an auditor's knock at your door.

References:1.Where Does Predictive Modeling Stand? Q&A With CMS Center for Program Integrity Deputy Director Ted Doolittle, Becker's Hospital Review, March 21, 2012
 
2. "Predictive Modeling Analysis of Medicare Claims." Publication. Department of Health and Human Services, Centers for Medicare and Medicaid Services, 2010. Web. 10 July 2012. 

3. LeGrand, Mary, and M. Bradford Henley. "Are You an E&M Outlier?" AAOS Now, Aug. 2009.

Natalie Prazen is a client services assistant and research associate at Karen Zupko & Associates. Karen Zupko is a seasoned senior advisor who has been helping physicians to navigate America's healthcare system since 1974. She is a member of the American Marketing Association and has served on the board of trustees of Chicago's Grant Hospital. You can contact her at kzupko@karenzupko.com.