As States Continue To Experiment In Medicaid, Look To Section 1115 Waiver Evaluations To Understand What Works
Flexibility and innovation have long been central tenets of Medicaid, the state-federal program that provides health care coverage to more than 77 million Americans. Through Section 1115 Medicaid demonstration waivers, states are allowed to make programmatic changes that are not otherwise permitted under federal law, provided the changes promote the goals of the Medicaid program and are budget neutral. Over the years, states have used 1115 waivers in many different ways, reflecting changing priorities of states and presidential administrations.
With this flexibility comes the requirement that 1115 waiver demonstrations be monitored and evaluated so that evidence on these experiments can inform future Medicaid policy. Yet, waiver evaluations have sometimes received short shrift, prompting some to raise concerns about the quality, timeliness, and usefulness of evaluation results. In recent years, the Centers for Medicare and Medicaid Services (CMS) has sought to strengthen waiver evaluations by providing guidance and technical assistance to states and their independent evaluators, including tools and guidance on evaluation design released last year.
Beyond the state-led evaluations required by CMS, numerous other studies have sought to understand the effects of 1115 waiver policies to inform future decision making. These include several studies supported by the Robert Wood Johnson Foundation’s (RWJF’s) Research in Transforming Health and Health Care Systems (RTHS) program, managed by AcademyHealth. The studies in this group make an important contribution to the evidence base by focusing on the implementation and impact of various 1115 waiver policies, including the opportunities and challenges they faced.
In this post, we highlight findings from RTHS research on three types of Section 1115 waiver policies: initiatives targeting justice-involved populations, work and community engagement requirements, and housing-related services. In doing so, we identify learnings that can help inform future research and evaluation and contribute to efforts by states and the federal government to inform future Medicaid policy making.
Justice-Involved Individuals: One-Size Outcome Measures Do Not Fit All
A small but growing number of states have sought Section 1115 demonstration waivers that include initiatives targeting justice-involved individuals, a population with complex health care and social needs. These waivers are among the tools available to states as they work to improve health, address disparities, and reduce recidivism among this population, particularly as individuals transition from jail or prison back into the community.
Researchers with the State Health Access Data Assistance Center at the University of Minnesota examined how three diverse states—Illinois, Texas, and Washington—approached the development, launch, and management of programs targeting justice-involved populations under Section 1115 waivers. In each state, these programs were a small component of a larger delivery system reform or substance use disorder waiver that included multiple initiatives and target populations. The programs targeting justice-involved populations varied across states but included things such as transitional care management for Medicaid enrollees exiting prison or jail, and case management for individuals with an opioid or substance use diagnosis who qualified for diversion into treatment outside of the criminal justice system.
Despite differences across the study states and their programs, the researchers identified several common factors that aided in successful program implementation. These included ongoing and frequent communication among partners, close coordination with law enforcement, and leveraging existing initiatives and provider infrastructures, rather than starting from scratch. Consistent challenges across states included aligning the goals of multiple sectors; data sharing and related privacy and legal issues; and hiring and training culturally competent staff, among others.
Study findings also highlight the pros and cons of carrying out programs for justice-involved populations within the framework and requirements of a Section 1115 waiver. Interviewees from the study states reported that waivers increased collaboration among stakeholders, supported new types of infrastructure that could be used to identify disparities, and expanded access to services that were otherwise inaccessible to the justice-involved population, among other benefits. However, they raised concerns about the ability of current waiver reporting and evaluation approaches to measure the success of initiatives focused on justice-involved populations. For example, while most states’ outcome measures are related to Medicaid costs and health care use, interviewees noted that indicators such as quality of life, retention in treatment, and equity should also be used to measure the success of justice-involved initiatives. Interviewees also noted that state-led evaluations for CMS tend to focus on indicators of success that can be calculated at the state level, whereas much of the activity and impacts associated with justice-involved initiatives occur at the local level. These findings suggest that a different or broader set of outcome measures may be needed in some cases to reflect the unique impacts of different types of waiver initiatives.
Work And Community Engagement Requirements: Assessing Stakeholder Experiences
Perhaps the highest profile use of Section 1115 waivers in recent years has been to institute work and community engagement requirements as a condition of Medicaid eligibility for certain populations. In February 2021, the Biden administration began the process of ending work requirements in Medicaid, leaving Arkansas as the only state where coverage terminations occurred as a result of failure to meet Medicaid work requirements. While Medicaid work requirements seem unlikely to get much traction in the near term, RTHS research on the experiences of Arkansas and New Hampshire—another early adopter—can help inform how other policies enacted under 1115 waivers are evaluated.
In early 2019, researchers with the Arkansas Center for Health Improvement interviewed 32 health care, educational, and community organizations across Arkansas that the state identified as sources of assistance for enrollees in meeting the work and community engagement requirement. Among these organizations, more than one-third reported having no knowledge or understanding of the requirement, and fewer than 10 organizations reported having the capacity to help enrollees with things such as calling the Medicaid helpline or assisting with exemption documentation. At the same time, very few organizations reported receiving requests for assistance from people in their community regarding the requirement. While these findings are based on a limited sample, they are notable given the extensive efforts by Arkansas state officials and their partners to communicate the requirement to both enrollees and stakeholders. Despite considerable outreach, awareness of the requirement was low among some organizations that were expected to assist enrollees, and few had the capacity to provide that assistance.
In New Hampshire, many community organizations were generally aware of their state’s work requirements but faced significant challenges preparing for implementation given the impact of the requirements on the population they serve. Drawing on stakeholder roundtables, interviews, and a survey, researchers with the University of New Hampshire found that community and health care organizations in New Hampshire instituted a range of policy, program, and infrastructure changes in anticipation of the state’s work requirement, which was suspended by the state and blocked by a federal court before any coverage terminations occurred. In the months prior to the policy’s suspension, community and health care organizations spent significant time and resources trying to identify which of their clients or patients would be subject to the requirement, developing and disseminating educational materials, and preparing for modifications to their information technology systems that would be needed once the work requirement took effect. Efforts by these organizations to understand and prepare for the requirement were complicated by other Medicaid policy changes occurring at the same time, including movement of the state’s Medicaid expansion population into managed care and the re-procurement of managed care contracts. Ultimately, New Hampshire’s work and community engagement requirement proved costly and confusing for community organizations, the researchers concluded, with many organizations realizing uncompensated costs that are not typically considered in the reporting and evaluations required by CMS.
These findings illustrate the key role that community organizations and other Medicaid stakeholders play in the implementation of 1115 waiver initiatives. Evaluations that consider the experiences of these organizations—including their expected versus actual roles and the costs they incur—can help inform a fuller picture of policy impact.
Housing And Medicaid: Implementation Research Investigates Work Across Sectors
Recognizing the relationship between housing and health, some states have used Section 1115 waivers to provide services that assist Medicaid enrollees with securing and maintaining housing. Among the target populations for these efforts have been individuals experiencing chronic homelessness, a group at higher risk for a range of adverse health outcomes relative to the general population.
To better understand the start-up implementation challenges associated with Medicaid housing support demonstrations, researchers with the Rutgers Center for State Health Policy examined the experiences of the first four states to receive CMS approval for these initiatives: California, Maryland, Washington, and Illinois. Demonstrations in these states targeted Medicaid enrollees with complex health needs who were homeless or at risk for homelessness and included pre-tenancy support, such as assistance locating and applying for housing, and tenancy-sustaining services, such as financial counseling and mediating tenant-landlord relations. These demonstrations also differed in many ways, including in aspects of their design, such as the extent to which states adopted a locally driven intergovernmental approach to carry out the demonstration versus engaging a third-party private contractor.
Drawing on interviews with 36 stakeholders in the four study states, the researchers identified several challenges to successful implementation of housing support demonstrations. These challenges included issues with housing supply, silos between health care and homeless service providers, contracting with and paying housing providers, and recruiting and retaining key workers. Despite these challenges, most of the study states made significant progress in program implementation, and the researchers identified many lessons learned. They also note that the limitations of housing support demonstrations suggest that federal policy makers should consider allowing states to more directly subsidize housing programs for those experiencing or at risk of homelessness as an optional Medicaid benefit.
Findings from this study exemplify the important role of implementation research in complex Section 1115 waiver demonstrations. The researchers identified challenges with housing support demonstrations that suggest numerous process measures for evaluators to consider, in addition to the key outcome of the degree to which people served by the program were housed. For example, what is the availability of housing in the area served? Were there barriers in communication between health care and homeless service providers? Were there challenges with contracting with or paying housing service providers, or recruiting/retaining key workers? Illuminating state responses to these kinds of issues, as well as exploring other barriers or successes, could be helpful to other states facing similar challenges.
Findings from these and other studies can help inform how waivers are evaluated in the future; they also underscore the value of continued collaboration and learning among states, evaluators, CMS, and others. In recent years, RWJF and AcademyHealth have contributed to this effort via a learning collaborative that resulted in a July 2020 field guide on real-world lessons in evaluation and design measurement. We encourage state and federal policy makers to continue to turn to the research to enact equitable policies that protect the millions of people who rely on Medicaid and who are facing some of the most significant health, economic, and social challenges of their lifetimes.