Methodology

<p class="htCenter"><strong>HCFA MEDICAID HCBS STUDY METHODOLOGY</strong></p> HCFA has contracted with The Lewin Group, and its subcontractors, the Urban Institute, the University of Minnesota, Mathematica Policy Research, Inc., and the MEDSTAT Group, to design and implement a study of the impact of Medicaid home and community based services programs on quality of life, quality of care, utilization and cost. The scope of the study will include both Medicaid home and community-based services waiver programs as well as other Medicaid-funded long-term care services. The research project will study the financing and delivery of services to older and younger people with disabilities in six states and the Medicaid financing and delivery of services for individuals with mental retardation and developmental disabilities (MR/DD) in six states. The study has two major components. During the first phase of the project, the Urban Institute developed case studies of the home and community-based services financing and delivery system for older and younger adults with disabilities in each of the six states and the University of Minnesota conducted case studies in six states for individuals with MR/DD. The A/D states selected were: Alabama, Kentucky, Maryland, Michigan, Washington and Wisconsin. The MR/DD selected were: Kansas, Indiana, Louisiana, New Jersey, Vermont, and Wyoming. In addition, The Lewin Group made a site visit to the Texas Star Plus program to explore this managed and capitated system for HCBS in the Houston area. Site visits were made to selected states to conduct unstructured in-person interviews with state officials, home care providers, trade associations and aging and disability advocacy groups. The case studies characterized key program variables such as eligibility criteria, service use, quality of care and care management controls. Aggregate level data were also collected on service use, costs, participant characteristics, and satisfaction, where available. Interviews were conducted between December 1999 and July 2000. The criteria for selecting states attempted to classify programs based on level of “maturity” or development of home and community-based services (HCBS). The rationale for classifying states in terms of their level of “maturity” related to their HCBS delivery system was that states and programs would vary on too many dimensions to be able to analyze the independent impact of design features. After gathering extensive data on program characteristics, we concluded that even if we were able to include all states in the study, it still would be difficult to examine individual aspects of programs. As a result, we opted to broadly characterize state’s systems through a ranking system used to select the study sites. As part of the second phase of the project, Mathematica Policy Research, Inc. (MPR), will conduct surveys of Medicaid home and community-based service users. The MEDSTAT Group will also obtain administrative data (Medicare and MSIS data) to research program costs in the selected sites. Lewin, Urban, and Minnesota will analyze survey and administrative data to examine data on service use, costs, participant characteristics, and satisfaction. Outcomes from each of the sites studied will be assessed by controlling for participant characteristics (e.g., age, disability level, caregiver availability) to determine the independent effect of each program on the outcomes. The features and structure of programs gathered through the process portion of the study will be used to compare programs with more favorable outcomes to those with lower ratings on outcomes to draw inferences on the promising program design features. This approach would treat program designs as complete packages and permit variation in more than one key dimension. Also as part of Phase II, we will examine the characteristics and care patterns that influence outcomes at an individual level. For example, does receiving more services produce a higher quality of life? Do people in residential settings have a lower quality of life than people living at home? And are people who live alone significantly less disabled than those living with others? The study will use a quasi-experimental design that gathers data from Medicaid recipients of both developed and developing HCBS systems. Information will be collected from several sources: site visits; surveys of recipients and proxies (where necessary) each within two groups of enrollees - elderly in the community and young disabled adults); interviews with state officials, advocacy groups, provider representatives and other key stakeholders; and secondary data sources, including MSIS data, and state administrative data. The analyses center on four major areas for developed programs relative to developing programs, as well as the influence of individual characteristics and care patterns: greater access to, choice among, and autonomy related to HCBS better outcomes (including fewer avoidable hospitalizations, lower death rates, higher levels of satisfaction with life and care, less unmet need) differences in patterns of care provided (fewer NF admissions and more discharges from NF to community, greater use of alternative residential facilities) differences in the cost of care provided (HCBS versus institutional, total health care costs, acute care spending versus LTC spending) The data for the study will come from several sources: utilization and spending data from the state Medicaid programs submitted to HCFA (MSIS) case studies of each site throughout the period of the study surveys of Medicaid HCBS recipients