CMS using analytics to catch fraud

Continued from Fraud Management page …

Farlen Halikman, Moore Stephens Lovelace CPAs

Farlen Halikman, Moore Stephens Lovelace CPAs

The software also compares data from the complaint line at the Medicare call center. Finally, CMS’ fraud system analyzes information from its Integrated Data Repository, which contains years of parts A and B claims, information about beneficiaries, and Part D drug data.

When the powerful analytics identify a suspect claim, it flags the document or documents with an alert. The alerts ignite the ZPICs, who then go after those providers – some 938 healthcare providers and facilities in 2014, the CMS report states.

Causing headaches for companies

Laura Wilson, operations manager for Van Halem Group LLC, said CMS’ push to halt fraudulent claims is being felt in the sector her firm helps clients navigate complex issues related to Medicare and Medicaid.

“Fraud in the Medicare realm is definitely causing headaches for companies in the durable medical equipment business,” Wilson said. “Unfortunately, it is affecting the ‘good guys’, those who have tried to do everything by the book.”

She suggests healthcare providers not take aggressive contractors personally.

“Something to keep in mind is that the [ZPIC] contractors are not singling out any provider, even though it may feel that way.  Rest assured that others are experiencing the same thing as you.”

The Van Halem Group educates clients to prepare them in case they get hit with an audit – whether it be a recovery audit, comprehensive error rate testing audit or a ZPIC audit, Wilson said. “The best way to go into an audit is with the knowledge you have the documentation needed to get you through.”

Improbable cause?

The problem lies somewhere between the alerts produced by CMS’ fraud analytics software and standards for probable cause, Halikman argues.

“With data analytics, just because you are an outlier doesn’t mean you are a thief,” he said. “ZPICs aren’t cops, they are contractors, not employees of the government. They get some money for successfully catching real fraud, if they can recoup money to the program, they get paid more.”

Halikman offers the following compliance tips to hospital administrators, doctor’s offices and other healthcare providers who bill Medicare:

  • The first rule is, have good documentation. If you have compliant documentation you are going to be OK,” Halikman says. “If you don’t, you are going to be in deep trouble. Unfortunately, doctors and long-term care providers don’t dedicate their lives to these care-giving roles because they have a burning desire for compliant documentation,” Halikman jokes. “But CMS problems usually stem from documentation issues.”
  • Know your PEPPER score. “You have to look and see your Program for Evaluating Payment Patterns Electronic Report (PEPPER) score,” Halikman said. The software tool developed by TMF Health Quality Institute (tmf.org) for CMS helps skilled nursing facilities assess their risk for improper Medicare payments. It uses CMS data analytics to compare you to your peers on a local, state and national level. The PEPPER score lets you know how close you are to the norm. “Every provider in the Medicare program should be downloading their PEPPER score off the CMS score card and reviewing it with their compliance officer,” Halikman says.
  • Have a very good electronic health record system. A good system will alert you when you are off the mark. For instance, when entering patient care or billing data into software on a mobile device, laptop or desktop computer, the screen one just “completed” does not move on to the next screen unless all the fields in the previous screen are filled in correctly, including with the right billing codes and other data.

mglass

Posted in Uncategorized.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>