Zone Program Integrity Contractors' Data Issues Hinder Effective Oversight

We found that workload data used by CMS to oversee the Zone Program Integrity Contractors (ZPIC) were not accurate or uniform. Inaccuracies and a lack of uniformity resulted from system issues in CMS's Analysis, Reporting, and Tracking System (CMS ARTS); ZPIC reporting errors; ZPICs' interpretations of workload definitions; and inconsistencies in requests for information reports.

ZPICs are replacing CMS's Program Safeguard Contractors (PSC) and will perform Medicare Parts A and B program integrity work in seven newly established geographical zones. CMS awarded the first two ZPIC contracts for Zones 4 and 7 on September 30, 2008, and both ZPICs were operational as of February 1, 2009. Our study was limited to the ZPICs for Zones 4 and 7 because they were the only ZPICs that had completed a full contract year at the time of our review. We collected and reviewed ZPICs' workload data related to investigations, case referrals, requests for information, and administrative actions from February 1 through October 31, 2009. Additionally, we reviewed ZPICs' performance evaluations and surveyed the ZPICs to identify any barriers they encountered in performing their program integrity activities.

One of our objectives was to describe the extent of ZPICs' program integrity activities during their first year of operation. However, the inaccuracies and lack of uniformity in the ZPICs' data prevented us from making a conclusive assessment of their activities. The lack of uniformity in ZPICs' reporting of data is similar to issues we identified more than 10 years ago in a review of Medicare Part A fraud units. Additionally, we found that ZPICs' performance evaluations contained few workload statistics and that data access issues affected ZPICs' ability to perform program integrity activities.

The issues we identified present a serious obstacle to CMS in effectively overseeing ZPIC operations. It is important that these issues be corrected so that CMS can analyze ZPICs' effectiveness in detecting and deterring fraud, waste, and abuse. It is also important for these issues to be corrected so that CMS can determine how well ZPICs are performing compared to other ZPICs and PSCs. Therefore, we recommend that CMS: (1) clarify the workload definitions in CMS ARTS to ensure that ZPICs' workload statistics are accurate and that ZPICs report their data uniformly, (2) improve oversight of ZPICs by performing a timely review of data in CMS ARTS for each ZPIC and across ZPICs to detect any anomalies in workload reporting, (3) utilize and report ZPIC workload statistics in ZPIC evaluations, and (4) ensure that ZPICs have access to all data necessary to effectively carry out their program integrity activities.

CMS concurred with some of our recommendations. CMS believes it has already complied with our first recommendation. CMS concurred with the first part of our second recommendation; however, it did not concur with the second part and stated that anomalies cannot be detected across ZPICs because of the differences in fraud landscapes between the ZPICs. We disagree and note that we detected anomalies in reporting between ZPICs, including differences in the way ZPICs were reporting their numbers of new investigations. CMS partially concurred with our third recommendation and stated it will consider including workload statistics for future evaluations, if appropriate. CMS concurred with our fourth recommendation.