What Burning Research Questions Do State Medicaid Directors Have?

With the current upheaval in the Medicaid program, what issues are top of mind for state Medicaid directors? At a recent meeting of the Medicaid Insights Colloquium, Hannah Maniates, director of Medicaid programming at the National Association of Medicaid Directors (NAMD) spoke about the research priorities that NAMD is hearing about from its members, including the impact of the reconciliation bill passed by Congress in July.

NAMD represents the 56 Medicaid directors from the states, five U.S. territories and Washington, D.C.

The Medicaid Insights Colloquium program is hosted by the Center for Health Systems Effectiveness at Oregon Health & Science University in Portland.

Maniates said that NAMD held a listening session recently during which Medicaid directors and senior agency staff identified several research priorities including mental health and substance use, One Big Beautiful Bill Act (OBBBA) implementations, how changes in eligibility may influence enrollment and outcomes, and benefit delivery trends, including around long-term care, pharmacy and managed care. 

Big focus on behavioral health

Medicaid agencies have pointed to the increases in behavioral health utilization they have seen coming out of the pandemic, Maniates said. “Given the context around state budgets being more constrained, Medicaid directors pointed to a number of areas where it would be helpful to have more research. First, what approaches can states take to manage budget pressures while maintaining access to behavioral health services? 

“States have questions about the most foundational evidence-based practices for mental health and substance use disorder,” she said. “What services are most critical to improving outcomes? Are there services where there's actually not a very good evidence base for their efficacy?” 

Medicaid directors also pointed to the overlap of behavioral health interventions, she explained. Many individuals are receiving multiple behavioral health interventions. How might those different interventions interact to influence outcomes, and how can they capture efficacy across those overlapping domains of care? For instance, if an individual in psychotherapy is also receiving a psychiatric medication and is also in a work program, how do we understand how those interventions interact to influence outcomes, and how can we capture outcomes across those different interventions? 

States also raised a number of questions around the best systems of care for young people with intellectual or developmental disabilities who have significant behavioral health challenges, Maniates said. “Medicaid directors said they're really interested in work around more coordinated systems of care for young people who have the most significant behavioral health needs.”

States also raised a number of questions around quality measures for mental health and substance use, she noted. “We know that a lot of process measures like emergency department visits and hospitalizations show up when someone's already experiencing fairly acute healthcare outcomes, so Medicaid directors said they're interested in measures that may indicate mental health and substance use challenges earlier, but then can be used to drive early intervention.”

Many behavioral health measures are tied to claims data, and policymakers are oftentimes interested in outcomes beyond utilization of healthcare. “They are interested in if someone has found employment, how their relationships are with their friends and family, their general life satisfaction,” Maniates said. “What are the ways to capture outcomes beyond healthcare utilization?”

Along a similar thread, states asked about patient-defined measures. What measures might reflect what matters to patients in terms of their recovery? 

States highlighted that providers have concerns around measurement, especially around behavioral health, so they may need to think creatively about ways to not just add new measures, but also manage the overall measurement portfolio. Are there measures that are being collected for other purposes that could be leveraged in the Medicaid program, and are there measures that don't work so well that could be sunset or replaced?

States raised a number of questions around Applied Behavioral Analysis (ABA), which is a service for autism spectrum disorder. Medicaid agencies are reporting very rapid increases in utilization and cost, raising questions around sustainability of those services, but also around program integrity, she said. “States asked what we know about which services are effective and what intensity of services? What is some of the clinical evidence around best practices for young people with autism spectrum disorder?” 

Medicaid directors also asked if there is a relationship between ownership structures of some of these ABA practices, such as private equity-backed practices, and some of these trends in cost and utilization? How can they watch out for any billing patterns that may be indicative of fraud, waste and abuse, while still maintaining access to critical services for members?

Research questions around OBBBA

Medicaid agencies pointed to a number of research questions that they're interested in as states implement OBBBA changes, especially the eligibility and enrollment policies, Maniates said. First, how will enrollment in Medicaid and CHIP change as a result of OBBA implementation? Will we see changes not just in caseload but also in case acuity? In the case mix of folks who are in the program, are there lessons learned from the Medicaid unwinding that can inform member outreach under some of the upcoming eligibility changes? 

States will need to implement cost sharing for expansion enrollees. They need to know about designing cost-sharing structures. Are there existing evaluations of cost sharing in Medicaid or in other programs?

States pointed to a number of questions around benefits and delivery systems, including pharmacy, Maniates explained. “The context here is that states are balancing the potential positive health impacts of some of the new high-cost and blockbuster drugs that we've seen on the market with the very real impacts on state budgets. GLP-1s like Ozempic and wegovy are a good example. What impacts are GLP-1s having on health outcomes and costs? Do we see GLP-1s reduce spending in the future on other healthcare services, and if so on what timeline should states expect to see those savings?”

States also raised questions around long-term care services. Over the past couple of decades, states have expanded access to home- and community-based services (HCBS) in their Medicaid programs, Maniates said. “States are interested in the impact of HCBS on Medicaid costs, and they ask specifically: Do we see an effect of HCBS on reduced nursing home or institutional care placement? Des HCBS act as a way to divert folks from being placed in nursing facilities or institutions?”

More research needed on managed care

States have a lot of questions on managed care. About two-thirds of Medicaid members are currently covered through managed care, so it is the predominant delivery model. “We know that the research on managed care has had variable findings,” Maniates added. “States are interested in more research on the impact of managed care on quality of care, health outcomes, and cost in the program. They are curious about what factors may be driving some of those variable findings.” 

State Medicaid directors also want to learn more about value-based care models, she said. Are there good examples of value-based care models in states that may be most effective for managing cost and improving outcomes in the program? 

Finally, Medicaid directors also pointed to questions around treatment adherence and no-show rates. “Some Medicaid directors said there's this popular idea that Medicaid patients are more likely to no-show or less likely to adhere to treatment plans,” she noted. “Is that true? Do we have empirical research on differences in no-show rates and treatment adherence across those covered by Medicaid vs. Medicare vs. marketplace? And if we do see differences, what are some of the factors that might be driving those differences?”