Methods To Prevent Fraud and Abuse Among Staff and Participants in the WIC Program Volume II, Site Visits E-FAN-01-012 December 2001 By Dev R. Chaudhari and Vicki Shaffer, Applied Techno-Management Systems, Inc., and Chris Logan, Abt Associates Inc. Abstract Summaries and highlights of site visits to eight States and two Indian Tribal Organizations are the basis of this report, which is part of a project to identify and assess methods used to detect and prevent fraud and abuse among staff or beneficiaries of the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC). It is the second of two related reports on the subject by the same authors. This report documents current practices in WIC Program monitoring activities and controls, and gives qualitative analyses of existing or proposed tools to assess the methods used by WIC managers to prevent and detect fraud. The report describes those basic operations and monitoring activities, and recommends existing or new tools that might be used to improve the integrity of WIC Programs. The primary purpose of the site visits was to find and document current practices in WIC program monitoring activities and controls to detect and prevent fraud and abuse by program participants and staff. For more information on this study, see Methods To Prevent Fraud and Abuse Among Staff and Participants in the WIC Program: Volume I, Final Report, E-FAN-01-011. This study was conducted by Applied Techno-Management Systems, Inc., and Abt Associates Inc. under a cooperative research contract with USDA’s Economic Research Service (ERS) Food Assistance and Nutrition Research Program (FANRP): contract number 53-3K06-9-10 (ERS project representative: Alex Majchrowicz). The views expressed are those of the authors and not necessarily those of ERS or USDA. CONTENTS Introduction/Backgroun iii State/ITO Site Visits Summary Reports 1 1. Arizona 2 2. California 10 3. Illinois 25 4. Kansas 37 5. Massachusetts 46 6. Tennessee 65 7. Texas 79 8. Virginia 93 9. Choctaw Nation ITO 101 10. Navajo Nation ITO 112 Introduction/Background The Economic Research Service (ERS) of the U. S. Department of Agriculture (USDA) contracted with the team of ATMS and Abt Associates to identify and evaluate methods that are used by various State or local agencies to detect and prevent fraud and abuse among staff or participants in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC). Once identified, best practices would be made available to all State and local agencies to prevent staff and participant fraud and help manage their WIC programs. The overall objectives of the ERS WIC study are to: (cid:1) Characterize basic operations and monitoring activities currently used by State or local WIC offices to identify WIC fraud and abuse among staff or participants (cid:1) Identify and evaluate the methods of detecting staff and participant fraud by reviewing existing tools and data analysis systems presently used by selected States/local and Indian Tribal Organizations (ITO) WIC agencies (cid:1) Perform qualitative analyses of existing or proposed tools to assess relative value/usefulness of methods (cid:1) Identify major legal, operational, political, and attitudinal limitations that may restrict adoption of tools to detect and control staff or participant fraud (cid:1) Recommend existing or new tools that are deemed to be effective best practices that State or local agencies may adopt to improve the integrity of their WIC Programs, while noting that those tools may not be the uniform best practice for all agencies (cid:1) Develop a compendium of best practices reflecting effective strategies, approaches and tools/techniques that are currently in use or can be used for WIC fraud and abuse detection/prevention. Such a document, once completed, can be used by various State and local agencies as a guide in supporting their efforts for fraud and abuse detection and prevention. In this document, the ATMS/Abt Associates team presents summaries and highlights of our site visits to selected State/local and ITO WIC agencies for the study. The primary goal of the visits was to find and document current practices in WIC Program monitoring activities and controls to detect and prevent fraud/abuse by Program participants and staff. Review of GAO and FNS Survey Data In an earlier report, we reviewed the General Accounting Office (GAO) and Food and Nutrition Service (FNS)/USDA surveys data in order to identify the data collection requirements during the planned site visits. The GAO survey, published in 1999 and FNS survey, published in 1998, iii primarily provide a broad nationwide profile of the current practices to prevent and detect fraud/abuse by WIC participants and staff. The GAO survey, conducted in 1998, was performed in response to concerns about the level of fraud and abuse in the WIC Program, coupled with the need to update the studies upon which USDA had relied for data on fraud and abuse. The survey is a good source of information about the types of fraud and abuse that are occurring, the basic preventive measures that are being taken by agencies, and the sanctions employed. The GAO survey provides statistical information, but does not provide detailed descriptions of how the controls operate. The GAO data were collected in 1998, before the implementation of the new WIC Program regulations concerning documentation of eligibility. The National State agency Program Integrity Profile, which was compiled by FNS in 1998, contains statistical information about the practices and policies of 77 State agencies and Indian Tribal Organizations. It includes basic information on policies that are geared toward preventing fraud and abuse by vendors, staff and participants. This information is largely quantitative, and offers no interpretation of the numbers. A second survey, the WIC Program Dual Participation Survey Summary was conducted by a FNS working group in 1998. It provides statistical information about methods employed by State agencies to prevent and identify fraud and abuse, as well as how agencies respond when fraud and abuse is identified. The working group also produced a document presenting model policies and procedures to detect, prevent, and resolve dual participation. Selection of State/Local/ITO Agencies for Site Visits In the report based on GAO and FNS surveys, a total of eight states and two ITO agencies were selected, in coordination with the FNS headquarters and regional offices, for site visits and associated data collection, in preparation for this report. The selection of the relatively effective State/ITO WIC agencies was based on criteria that included key considerations such as: (cid:1) Representative State size (large, medium, small) (cid:1) Adequate geographical and regional representation (cid:1) Relevant extant laws and policies (cid:1) Availability of data related to fraud/abuse (cid:1) Existence of participant and staff fraud/abuse controls (cid:1) Enforcement actions and impacts (cid:1) Management assessment of the effectiveness and feasibility of controls (cid:1) Range/degree of challenges faced by the State/ITO agency (cid:1) Interest and willingness to participate in the study The selected agencies for site visits were located in: Arizona, California, Illinois, Kansas, Massachusetts, Tennessee, Texas, Virginia, the Choctaw Nation and the Navajo Nation. Virginia was selected as an alternate to take the place of Maryland, which FNS indicated was unable to participate. The participating states and ITOs are considered to be among the leaders in fighting fraud. There are also numerous other agencies with similarly strong practices. They are a diverse group in terms of size, location and management information systems (MIS) development. This diversity iv allowed the study team to explore what are the current best practices in the context of the existing program environment, including how agencies deal with resource limitations. The selection of the local WIC agencies within each State was coordinated with each individual State WIC agency using the criteria similar to that used for State selection. Two local WIC agencies were selected for site visit and data collection in each of the states with the exception of Tennessee and Arizona (where a single local agency was selected). Fourteen local agencies were visited to review and study the WIC Program implementation and operations at the “street level.” The ITO visits included both central administration and local clinic operations. Overall Conceptual Framework for Site Visits and Data Collection In our data collection and analysis efforts, we focused on the vulnerabilities to fraud/abuse in the certification, issuance, transaction and redemption-related processes by participants and program staff. The data collection and analysis focused on the following: (cid:1) What documentation of income or adjunct eligibility do local agencies obtain from WIC applicants? (cid:1) How do local agencies verify or validate documentation provided by applicants? (cid:1) What proof of residency do local agencies accept? What independent checks on residency do they perform? (cid:1) What checks on dual participation within State and across States do they perform? (cid:1) What identity verification or other controls are used to prevent and detect “phantom” cases (benefits issued to nonexistent participants)? (cid:1) Do local agencies require reporting of changes in eligibility? What are the requirements, and how are they enforced? Do agencies obtain independent information on continuing eligibility (e.g. through periodic data matches with Medicaid or other programs)? (cid:1) What procedures, computer system features, and security measures are used to prevent and detect fraud and abuse in the issuance process? (cid:1) What do State and local agencies do to follow up on possible recipient or staff involvement when vendor fraud is detected? (cid:1) What procedures in recruiting, hiring, training, and managing staff do State and local agencies use to prevent and detect participant and staff fraud? What kinds of staff behavior or performance indicators are used to trigger audits or investigations, and how are these inquiries conducted? (cid:1) What evidence or information, if any, is available from WIC administrative records to document the cost and effectiveness of controls on participant and staff fraud? v (cid:1) What are the legal, regulatory, and/or operational challenges to or constraints on wider implementation of current or promising controls? In brief, our approach to addressing these questions and meeting ERS’ requirements consists of the following: (cid:1) Integrating data from existing GAO and FNS surveys of State and local WIC agencies into a systematic description of current fraud control strategies. (cid:1) In eight states and two ITOs with innovative or well-regarded fraud control systems, we plan to conduct reviews of procedures, computerized applications and databases, noncomputerized records, and data on the use, effectiveness and results of the controls. (cid:1) Descriptive analysis of the patterns in the controls used against participant and staff fraud among State and local agencies. (cid:1) An assessment of the effectiveness, resource-intensiveness, and operational feasibility of current and proposed controls. (cid:1) Synthesis of the current practices and recommendations for FNS, State and local agency actions to improve the integrity of the WIC Program. Site Visits for Data Collection At the State level, the data collection efforts drew on a number of data sources including inputs and perspective of key staff related to program operations, use of management and information systems, data on fraud/abuse practices and investigations, and use/value of current and planned enforcement actions. These visits also provided the ATMS/Abt Associates team access to relevant data, documentation and technical expertise needed to understand and assess the usefulness of automated systems for fraud/abuse detection and prevention. In addition, available data or opinions on the costs and effectiveness of controls were assessed during these visits. The local agency site visits provided the opportunity to understand and evaluate how the controls operate at the “street level.” In the WIC Program, the diversity of organizations serving as local agencies makes their perspective particularly important, even though it is difficult to generalize. For this study, the data collection in local WIC agencies included interviews with agency managers and line workers, observation of interviews and use of automated applications, and reviews of documentation used to prevent and detect fraud. The data collection effort during the site visits was aimed at the following: (cid:1) Get detailed information on procedures or systems through interviews, observation and review of documentation. (cid:1) Understand how controls are implemented and the steps taken to make sure that they are sustainable and efficient. (cid:1) Get and review records of control-related activity and outcomes. vi (cid:1) Understand the context that gives rise to fraud and that has made it possible (or challenging) to implement the controls where they have been successful, so that recommendations for other States can address the conditions that may favor or impede implementation elsewhere. (cid:1) Review sensitive issues of how controls fit into WIC agencies’ mission, culture and political environment. (cid:1) Review and assessment of current or proposed controls by asking State and local WIC agency staff to critique them on the basis of their knowledge and experience. The site visits included a combination of semi-structured interviews, reviews of documentation, and observation of key procedures or computer applications. To the extent possible (within time limits, security requirements, and confidentiality restrictions), the data collectors also obtained copies of relevant documentation for computer systems, procedures, and other useful documents such as consent forms and notices to participants. The following section of this report provides a summary of our site visits and data collection efforts for each of the State and ITO WIC agencies visited. For each State/ITO WIC agency, it provides: (cid:1) Background information covering agency organization/staffing and program operations overview. (cid:1) Overview of program operations and processes with a particular focus on program monitoring activities and controls related to major program functions and activities. (cid:1) Summary of site visit results highlighting key practices and their effectiveness in program administration for fraud/abuse control. The information in these reports, together with the FNS and GAO data on broad patterns of fraud and abuse prevention and detection practices, are synthesized in the study’s final report. vii Site Visit Summary Report This section presents an overview of the results of our site visits and associated data collection efforts in each of the eight States and ITOs identified earlier. The information presented herein is subject to verification by individual State/ITO WIC agency through subsequent coordination for any material clarifications or changes. These results are presented separately for each State/ITO WIC agency in the following order: 1. Arizona 2. California 3. Illinois 4. Kansas 5. Massachusetts 6. Tennessee 7. Texas 8. Virginia 9. Choctaw Nation ITO 10. Navajo Nation ITO Subsequent to our site visits, ad hoc feedb ack from these State/ITO WIC agencies was obtained to determine the nature and extent of the use of risk indicators to detect fraud and abuse by WIC Program staff or participants. Our primary purpose for obtaining this feedback was: to identify which, if any, of the listed indicators are used; to find out whether their use is routine or on an ad hoc basis; to understand the usefulness and/or limitations of these indicators; to identify specific impediments in the use of these indicators; and to gather information on any other indicators that are currently in use. Since most of t he agencies visited use onl y a few of the indicators, and provided v ery brief responses, Exhibit 1–1 provides an effective means for summarizing the information provided by these a gencies. S ome of t he a gencies not l isted in this chart did not provide the requisite information. As can be seen from this exhibit, few agencies make routine use of these indicators, and key impediments to routine use inc lude: c urrent s ystem limita tions; ne ed f or s ystem enhancements; inadequacy of staff resources; and need for data sharing among others. It should be pointed out that the agencies that use t he indicators find them gene rally useful, and th ey would like to deploy or increase their use as existing systems are changed or enhanced, and new systems are developed. [Exhibit 1-1 is found at the end of the report.] 1 1. Arizona Site Visit Summary Report I. Background The State a gency describes Arizona as the s econd fastest growing state in the nation, with a population of about 4 million. Approx imately 60 percent of the participants are Hispanic, and only about half sp eak English, so ret ention of bilingual staff is particula rly important in the clinics. Approximately 75 percent of the families served by WIC in Arizona have one or more working parents. The State agency reports that USDA’s definition of the migrant family does not fit Arizona’s farm-working families in which the mother and children remain stationary while the father moves around following work. The Arizona WIC program also serves a large number of families of transient construction workers and military families. There are many large military bases, some of which have WIC clinics on base. The Arizona WIC Program rec eives $50 million in food funds , $20 million in Nutritional Services and Administration (NSA) funds and $25 million in rebates. Three local agencies have applicants on waiting lists. The State agency estimates that fewer than 1,000 are unserved due to lack of funds. In agencies where this is a problem, a priority system is in place to determine who is at highest risk. For those who must be denied benefits, referrals to other programs such as the Commodity Supplemental Food Program (CSFP) are made. In Arizona, CSFP is administered b y the W IC Program. WIC Program Organization The W IC P rogram in Arizona serves approximately 125,000 particip ants each year. W IC services are provided by 14 county health departments, 3 community health centers, and 1 ITO. The county health departments have intergovernmental agreements which are approved by the county board of supervisors. The community health centers have contractual agreements with the State agency that are re-evaluated every five years, and the ITO (Cocopah) works with the State agency rather than functioning independently. The State a gency in Ariz ona was visited durin g the w eek of J uly 24, 2000. The local agency visited was the Yavapai County Health Department in Prescott, AZ. At the State a gency l evel, the W IC Program falls unde r the Communit y & Family Health Services, Office of Nutrition Services. The nutritional assistance programs manager oversees the WIC Program. The WIC Program director oversees the local agency and the clinics which fall under it. Certification is the responsibility of the community nutrition worker (CNW). The CNW is a paraprofessional who is required to complete competency-based training. CNW’s must also fulfil an annual r equirement of 48 hours of continuin g training. Nutritionists are r eserved for high risk clients. The Arizona W IC Program has a cont ractual agreement with PDA Software Services, Inc. (PDA) to perform data entry and data processing, banking services, reconciliations, and the production of reports which are used to detect and prevent fraud and abuse (i.e. dual participation reports). Only PDA has direct access to the participant database; State and local WIC staff can only view MIS data via printed reports. PDA sends agencies a complete list of all participants 2