The Role of Medicaid Waiver Policy Design in Home and Community-Based Services Use Among Adults with Intellectual and Developmental Disabilities

Medicaid waivers provide states with flexibility to offer specialized benefits to priority populations. They are increasingly used to address some of the most complex health and social policy challenges in the United States, affording states wide latitude in design and administration, resulting in a varied national policy landscape with limited evidence-based guidance. This dissertation focuses on Medicaid 1915(c) home and community-based services (HCBS) waivers, the primary mechanism through which states provide long-term services and supports to older adults and people with disabilities. Existing research has largely relied on binary measures of waiver adoption or single-dimension proxies for state policy generosity, obscuring the multidimensional trade-offs states make in waiver policy design and limiting the ability to provide actionable policy guidance. People with intellectual and developmental disabilities (I/DD) have particular policy significance in this context, accounting for nearly three-quarters of both national Medicaid 1915(c) HCBS waiver spending and estimated unmet need for Medicaid-funded HCBS waiver services nationwide, yet the population remains critically understudied. This dissertation addresses these gaps by developing a novel Medicaid 1915(c) HCBS waiver policy dataset and linking it to individual-level Medicaid claims to examine the role of state waiver policy design in shaping waiver enrollment and use of waiver services among adults with I/DD, providing some of the first nationally representative evidence on how specific policy choices shape waiver access and HCBS use for this population.

This dissertation first constructs a novel Medicaid HCBS waiver policy dataset for people with I/DD by systematically coding the content of publicly available Medicaid 1915(c) HCBS waiver documents across 43 states and uses cluster analysis to identify a typology of three distinct state policy packages: states that prioritized broader eligibility and enrollment capacity with narrower benefit coverage and lower average spending allocated per waiver participant; states that allocated the highest average spending per waiver participant with broad benefit coverage and narrow eligibility and enrollment capacity; and states that optimized the scope of eligibility and benefit coverage with moderate spending per waiver participant, often through multiple tiered or targeted waivers. This typology captures deliberate and consequential trade-offs states make in waiver policy design that cannot be characterized by any single dimension.

The dissertation then links this novel policy dataset to individual-level Medicaid T-MSIS Analytic Files (TAF) data for adults with I/DD ages 18–64 to examine waiver enrollment across 38 states and the District of Columbia in 2019. States that prioritized broad scope of eligibility had substantially higher Medicaid 1915(c) waiver participation rates among adults with I/DD enrolled in Medicaid than other policy package types, indicating that the typology has meaningful explanatory power for enrollment. Among ancillary policy choices, which facilitate or enhance implementation of these policy packages, residing in states with no reported waiting list or unmet need for waiver services and higher hourly respite reimbursement were associated with higher enrollment probability, while the ability to pay immediate family members without limits was associated with lower enrollment. There was also meaningful variation in the association between state policy choices and enrollment probability across racial and ethnic groups. Most notably, participant direction opportunities enabling waiver participants to serve as the employer of record were associated with a 14.88 percentage point higher enrollment probability for Hispanic and Latino adults with I/DD but not for other racial and ethnic groups.

Finally, the dissertation extends this analysis to examine what services adults with I/DD enrolled in Medicaid 1915(c) HCBS waivers actually receive, linking the policy dataset to individual-level Medicaid TAF data for adults enrolled in waivers across 33 states. Among adults with identified Medicaid HCBS waiver enrollment and use of services, states that prioritized broader scope of benefit coverage were associated with the highest service breadth and annual spending, while states that prioritized broader scope of eligibility were associated with lower service breadth and spending alongside higher enrollment rates. States that optimized both eligibility and benefit coverage—often through multiple tiered or targeted waivers—did not differ significantly from states prioritizing benefit coverage on either service breadth or annual spending, despite the latter having substantially higher projected average spending per waiver participant and lower enrollment capacity, suggesting that how states structure their waiver policy design may matter as much as the overall level of generosity in shaping the intensity of services received by enrolled adults. The two models of participant direction were associated with fundamentally different service profiles, with budget authority linked to higher use of community participation and employment services and employer authority linked to lower service breadth and spending. Racial and ethnic disparities in service use were substantial and persistent. Most strikingly, Hispanic and Latino adults with I/DD enrolled in waivers had mean annual HCBS spending of $28,432, roughly two-thirds of the overall mean, despite comparable service breadth, suggesting that securing a waiver slot is necessary but not sufficient for equitable access to the full range of HCBS waiver services.

Collectively, these findings demonstrate that understanding and improving access to Medicaid HCBS for adults with I/DD requires attention to the full structure of state waiver policy design, not as a collection of isolated choices but as a multidimensional system whose consequences differ meaningfully across states and populations. These findings are particularly timely as significant reductions in federal Medicaid funding are already forcing states to make difficult choices about whether to modify eligibility and enrollment capacity, benefit coverage, or spending allocations, which are precisely the trade-offs this dissertation aims to make legible.