By 2023, it is expected that the number of VHA enrollees aged 65 and over will increase from 4.1 million to 4.7 million. To meet the growing demand for long-term care services, VA expanded its home and community-based services (HCBS) through the Millennium Health Care and Benefits Act. These expansion efforts were based on the premise that HCBS provide care in Veterans' setting of choice for a lower cost than in institutional settings and with comparable outcomes. However, there is sparse evidence about the ability of HCBS to reduce long-term care costs in VA or in other settings. Further, previous VA evidence on HCBS expansion has been unable to: 1) look at a time horizon over which cost savings are likely to be realized, 2) identify a well-matched control group; or 3) employ robust methods that enable causal assessments of the impact of HCBS on subsequent outcomes. Recent linkages between the Health and Retirement Study (HRS) and VA and Medicare data provide a unique opportunity overcome these shortcomings of the existing evidence.
The objectives of this study are to use the linked HRS, VA, and Medicare datasets to assess the feasibility of comparing long-term care utilization and quality outcomes for Veterans who were exposed to HCBS expansion efforts to outcomes for non-Veteran Medicare only users who were not exposed to HCBS expansion. We will: 1) compare the functional status and other characteristics of VHA users and non-Veteran Medicare groups; 2) identify the best methods to match the two groups on observable characteristics; 3) describe trends in quality of life, nursing home use, and formal and informal home care service use for these groups pre- and post- HCBS expansion; and 4) identify the most appropriate methods to determine the causal effects of HCBS expansion on each outcome.
We will conduct a longitudinal cohort analysis using HRS linkages to VA and Medicare data. We will use health and functional status data from the HRS to identify a cohort of Medicare users with comparable long-term care needs to the Veteran cohort. We will conduct descriptive and multivariate analyses comparing the long-term care need trajectories of these two groups from 1992 to 2012. This analysis will focus on identifying the most appropriate methods to match the two cohorts. Next, we will use a difference-in-differences approach to assess the quality of life, probability of nursing home entry, and use of formal and informal home care services for these two groups before and after the Millennium Act.
The anticipated result will be an assessment of the feasibility of using a non-Veteran Medicare cohort as a control group for Veterans who experienced HCBS expansion. We hope to determine the most appropriate methods to match the VA HRS cohort to a non-Veteran group of Medicare users who were not exposed to HCBS expansion. We will produce descriptive and multivariate analyses that show trends in self-reported long-term care usage and quality of care from the HRS.
Strengthening this evidence base will help VA understand Veteran and system-level effects of HCBS expansion and achieve an appropriate level of HCBS investment. These analyses will inform a future IIR that will use the linked HRS-VA and Medicare administrative data to determine the effect of HCBS expansion on Veteran and caregiver outcomes and VA system-level costs. This will allow us to assess whether HCBS expansion has been cost-effective from a VA perspective. The results of this pilot project will be relevant to Veterans, their caregivers, and VA policy makers involved in allocating long-term care funding.
External Links for this Project
Grant Number: I21HX002420-01
None at this time.