HHSC Response to OIG Report

TEXAS HEALTH AND HUMAN SERVICES COMMISSION

P. O. Box 13247
Austin, Texas 78711

4900 North Lamar,
Austin, Texas 78751 April 18, 2007

We recognize that the Office of Inspector General (OIG) was given a challenging task to
complete a review within about a month of an eight-year-old technology project and a procurement process that spanned more than a year. HHSC believes that proceeding with the OIG’s recommendation for an independent audit of the project would provide the public and state leaders with a higher level of confidence in the performance of the state’s new eligibility software and the plans for modernizing the state’s entire eligibility system.

One key fact often gets lost in the detailed discussions about specific elements of the new software system, known as the Texas Integrated Eligibility Redesign System (TIERS). The new system works. It already provides benefits to more than 250,000 clients each month, and the system’s accuracy has undergone several independent reviews:

• The state’s independent verification and validation contractor, Science Applications International Corporation (SAIC), directed side-by-side tests of TIERS and the legacy System for Application, Verification, Eligibility, Referral and Reporting (SAVERR) and found that the systems returned identical benefits calculations. This test involved having state staff complete cases in both systems. Both systems returned the same results.

• The State Auditor’s Office (SAO) retained KPMG to perform the federal compliance portion of the Statewide Single Audit for fiscal year 2006. The analysis noted concerns about incomplete documentation in some case files but did not find that TIERS performed any incorrect benefit calculations. HHSC is providing additional training to caseworkers about the importance of using the comments feature in TIERS to ensure that documentation is complete. After the SAO report, HHSC verified that documentation and recalculated the benefits for those cases. In each case, the original benefit calculations were found to be correct.

• Clifton Gunderson, an independent audit firm, conducted an IT application and security audit of TIERS. The August 21, 2006, report states, "Overall, our testing of TIERS application code (Java files) indicated that TIERS provides adequate processing and output controls to ensure data integrity."

HHSC is concerned about the integrity of OIG’s analysis of food stamp cases in TIERS. During the OIG test, the state workers entering the cases into TIERS and SAVERR were provided with different sets of instructions about how to process part of the case. This led to errors in both the TIERS and SAVERR benefit results. In addition, the OIG’s findings about the performance of SAVERR and TIERS are based on survey of 153 caseworkers rather than a true analysis of each system. 2

The OIG report is correct in noting that it takes more time to process an initial application in TIERS as compared to SAVERR. This is because the TIERS process is by design a more rigorous and complete look at the household. TIERS collects twice as many data elements as SAVERR. This saves time in the long run, strengthens fraud prevention, and provides a higher level of confidence in the state’s ability to accurately determine a family’s eligibility for state assistance. TIERS is simply a better and more complete system than the antiquated one it is replacing. For example:

• Because SAVERR collects less data on the household, a husband and wife could each seek benefits separately in two different state eligibility offices. There is no upfront safeguard in SAVERR to prevent the household from receiving two sets of food stamp benefits. TIERS would recognize and prevent this situation from occurring. With SAVERR, the state would have to pursue a fraud case and attempt to recover the benefits after the fact.

• It takes far less time to complete recertifications or other case actions in TIERS as compared to SAVERR. This is significant because 80 percent of case actions are recertifications.

Essentially, the state spends more time on the front-end processing of the application in TIERS
in exchange for far greater time savings in the future.

• TIERS automatically and uniformly applies all program policies including household composition and budget calculations. SAVERR relies on the worker applying policy at their discretion.

HHSC has committed to addressing any federal issues with TIERS before converting additional areas of the state to the new system. We have worked closely with federal oversight agencies since the initial planning phase of TIERS in 1999 and will continue to do so. The OIG’s claim that full TIERS implementation would result in billions of dollars in questioned costs is misleading.

The State Auditor’s Office questioned $78,438 in TIERS benefits based largely on insufficient documentation in the case files. As noted previously, HHSC verified the documentation, recalculated the benefits, and found that the original benefit calculations were correct.

The OIG report notes seven events over the eight years of TIERS development and implementation and the state’s efforts to use call centers that it defines as "critical events." HHSC would like to provide additional clarifying information about each of these events.

1. The removal of the original data warehousing requirement from the initial TIERS contract. HHSC Response: TIERS collects and stores far more information than SAVERR. In fact, the system is built on an Oracle database and provides the state with vastly improved data management capabilities. TIERS is a real-time, client-focused system unlike SAVERR. In SAVERR, a client who has food stamps and Medicaid has two case files, with two separate entries for addresses and other information associated with each case. TIERS collects and stores information about each client in one place, and allows workers to make address changes for a client’s food stamps, Medicaid and TANF benefits with one entry. It should be noted that a data warehouse environment would not meet OIG requirements for investigations. In a data warehouse, selected production data is modified to best support a reporting environment. OIG needs the data as it was entered in the system. Therefore their source must be the TIERS production database and not any associated data warehouse. HHSC has worked with OIG to address its request for additional reports that pull case information out of TIERS to assist with OIG investigations. The agency has attempted to explain that OIG processes may need to be adjusted to reflect the more comprehensive information available in TIERS. The OIG, however, has requested that HHSC produce exact duplicates of the SAVERR reports in TIERS. On April 2, HHSC added new capabilities to TIERS to accomplish this task. OIG has requested that the reports include a time stamp in addition to the current date stamp, and HHSC is in the process of completing that change.

2. The commitment of the state authorities to the TIERS application without consideration of the possibility of other alternative technical solutions. HHSC Response: The Legislature directed the development of TIERS in 1999 based on a thorough analysis that concluded SAVERR would soon be unable to meet the state’s needs. Since 1999, the state had invested $279 million in the development of TIERS and piloted the software in two counties. Nothing about the transition to an integrated elibility system gave HHSC reason to question the Legislature’s 1999 decision. HHSC did conduct an analysis of the requirements of the new integrated eligibility model and confirmed that SAVERR was incapable of supporting the goals of the project. HHSC also determined that using a state system, rather than vendor system, provided the state with greater flexibility for the long-term of the project. Using a vendor system makes it more difficult and costly for the state to change vendors or the project’s direction at a later date. Based on these factors, HHSC required that vendor proposals incorporate the use of the TIERS source database and its rules engine in their proposals. Vendors were not required to use any other element of TIERS, including the data collection functions.

3. The decision to privatize did not use all the relevant data that was available. HHSC Response: HHSC carefully evaluated both state-run and outsourced options for modernizing the state’s eligibility system. Either choice represented a significant transition for one of the state’s largest public systems. The state already outsourced many of the functions included in the new system, including CHIP call centers and processing, maintenance of key computer systems, and the enrollment of clients into managed care plans. An analysis found that rolling those services into one contract and transitioning additional administrative services to a vendor would save taxpayers $210 million more than a state-run system over a five-year period. HHSC’s business case was based on data from the legacy SAVERR system. However, the relative savings between a state-operated system and a vendor-operated system would have been the same regardless of which system’s data was used.

4. Although state authorities initially used the statistically valid SAO scoring methodology, a different methodology was then utilized in the process of awarding the IEES contract. HHSC Response: HHSC has acknowledged that the two leading bidders for project, Accenture and IBM, both submitted strong proposals and that their scores were very close. The OIG report fails to note that the state’s ultimate decision to award the contract to Accenture was based on the lower cost of that proposal. Accenture’s final offer was more than $50 million less than IBM’s offer. In addition, it should be noted that IBM contested the contract award in three courts, and each court rejected the challenge of HHSC’s procurement process.

5. State authorities released Deloitte from correction of numerous system deficiencies resulting in a failure to enforce original, fundamental requirements of the initial TIERS contract. HHSC Response: There is no way to transition a software project from one vendor to another without transferring any defects. All software has defects. Of the 579 defects mentioned in the OIG report, the vast majority were duplicate requests, training issues or other issues that were not true defects. HHSC did document 77 defects open in TIERS at the time of the transition and any functions that Deliotte did not deliver and negotiated an amendment to recover the costs associated with those defects from Deloitte. HHSC then negotiated contract amendments with Accenture to cover the costs of addressing six of those defects and to make additional system improvements and to comply with changes in state and federal law, such as changes to the federal poverty level guidelines.

6. State authorities released Accenture from a fundamental contract requirement immediately before going live with IEES, by agreeing to the outsourcing of a core competency requiring program policy knowledge. HHSC Response: HHSC has acknowledged that it proceeded with the pilot launch despite a known problem with interfaces between TIERS and a vendor system known as MAXe. HHSC worked with the vendor to assess the feasibility of moving forward with an interim solution that involved the manual entry of some data by the vendor at no cost to the state. This issue, coupled with additional system and training issues, led HHSC to suspend the pilot after six weeks of operation. While the pilot was not successful, the agency gained a great deal of valuable information and experience during its operations. The agency’s commitment from the project start was to carefully evaluate the results every step of the way and take appropriate actions. We did.

7. The absence of effective management of project and contract oversight. HHSC Response: The OIG report faults the agency for using expertise from both program and information technology staff to lead the project. Given the complexities of the project, the agency must pull from the expertise in all relevant areas – program, information technology, legal and finance. The leaders of these areas meet regularly to discuss the project’s performance and evaluate next steps. The Health and Human Services Executive Commissioner makes final decisions on key project activities with the advice of staff from each of these areas.