Health Policy$ense

Debating Medicaid Rules for Mental Health Care

"IMD exclusion" under scrutiny

In a recent Scattergood Foundation report, LDI Senior Fellow Dominic Sisti and I tackle the curious case of the “institutions for mental diseases” (IMD) exclusion in Medicaid. For non-elderly adults, the national IMD exclusion prevents Medicaid from paying for inpatient care in institutions with more than 16 beds that primarily provide care for persons with “mental diseases” other than dementia or intellectual disabilities. In the wake of the opioid epidemic, mass incarceration, and shortages of behavioral health care capacity, the IMD exclusion has come under renewed scrutiny.

To better inform the discussion, the Scattergood Foundation engaged Jennifer Mathis of the Bazelon Center for Mental Health Law to defend the IMD exclusion, while Sisti and I made the case for repealing it. The resulting paper is part of an ongoing series devoted to contemporary issues in behavioral health, which includes treatment for both mental illnesses and substance abuse disorders.

As with many issues in health policy, the current controversy has deep historical roots.  

The IMD exclusion is an original pillar of Medicaid
Behavioral health care has historically been a state prerogative. Even when Lyndon Johnson’s Great Society expanded the federal government’s role in the social safety net, there was considerable effort to avoid federalizing behavioral health. At the time of Medicaid’s passage in 1965, legislators were aware that federal funding for institutional care could diminish state interest in behavioral health, stunt the development of community-based care, and incentivize long-term institutionalization of people who could be treated in the community.

In response, the IMD exclusion was built into Medicaid. It reflects a skepticism of institutionalization based on civil rights concerns and a strategic decision to favor community-based treatment.

As Medicaid’s role has grown, a necessary question to ask is, has the IMD exclusion aged gracefully?

Today, Medicaid is the most important safety net program for persons with behavioral health conditions. In 2015, despite covering only 14% of the adult population, Medicaid provide insurance for over a quarter of adults with serious mental illness (SMI) and nearly a fifth of adults with substance use disorders (SUD). The program covers a broad range of services, including individual and group therapy, partial hospitalization, case management, detoxification, and supportive housing services. As Medicaid’s role has grown, a necessary question to ask is, has the IMD exclusion aged gracefully?

On the one hand, the original vision of state-based community treatment is now reality. Nearly all behavioral health care has shifted to community-based settings since the 1960s. In 1955, there were over 500,000 state hospital beds for patients with mental illness--a rate of 337 per 100,000. In 2018, the number of state hospital beds has dropped to 11.7 per 100,000. The IMD exclusion is a part of the story of national deinstitutionalization. On the other hand, most experts agree that America’s behavioral health system is in crisis. Some say the IMD rule is contributing to the problem.  

The contemporary debate
Mathis argues that the IMD exclusion has fostered the development of cost-effective community services, and protects a public mental health system that directs investment towards the least coercive setting. She refutes the claim that the exclusion exacerbates inpatient bed shortages or is discriminatory. While some patients struggle to access inpatient care, Mathis contends that those gaps stem from a failure to adequately fund community-based services that reduce the need for hospitalizations in the first place.  

She maintains that the correct approach to increasing access to inpatient beds is not to repeal the IMD exclusion, but rather, to expand cost-effective programs like assertive community treatment and supportive housing. The upshot of this argument is that repealing the IMD exclusion would serve to divert funding from already strained community-based services, which would worsen a national crisis.   

Sisti and I agree with much of what Mathis has to say. Most behavioral health should be provided outside of hospitals and seamlessly integrated with wrap-around social services. However, we disagree that the IMD exclusion is essential for protecting community-based care. In the limited cases where the IMD rule has been waived, there was no observable shift away from community-based care. But most importantly, we argue that even in places where community-based care is well-financed and robust, the rule fails to pass basic ethical tests.

Is the IMD exclusion unethical? On mental health parity and “institutionalization”
Even if the community psychiatry system were sufficiently funded, a proportion of seriously ill people will still require inpatient care. Given that reality, we argue that the IMD exclusion violates the concept of mental health parity, which dictates that mental and physical health are in the same “ontological realm”; there is no physical health without mental health. We doubt that the public would accept similar exclusions for physical health, such limitations on access to coronary artery bypass grafting based on the size of a hospital. 

Furthermore, we believe that policymakers should take a wider view of “institutionalization” when considering behavioral health policy. Large scale institutionalization of persons with substance abuse disorders and severe mental illness already occurs in state prisons. The IMD exclusion has failed to prevent—and may have even worsened—this type of non-therapeutic warehousing.

Looking ahead
Increasingly, exceptions to the IMD exclusion rule have been made, and calls for IMD reform continue. In April 2016, CMS issued a new managed care rule that allows states to receive federal funds for capitation payments to managed care plans that cover IMD services for up to 15 days a month. States can also direct federal funds to IMDs through Medicaid Disproportionate Share Hospital (DSH) payments and Section 1115 Waivers. As of June 2018, 24 states have either approved or pending waiver requests related to IMDs. In April 2018, the House Energy and Commerce Subcommittee on Health considered legislation that would partially repeal the IMD exclusion as part of a series of bills to address the opioid crisis. The Scattergood Foundation report provides a strong foundation for understanding these debates.


Aaron Glickman, BA, is a policy analyst in the Department of Medical Ethics and Health Policy, Perelman School of Medicine, and in the Leonard Davis Institute for Health Economics, University of Pennsylvania.

Dominic A. Sisti, PhD, is Director of the Scattergood Program for Applied Ethics in Behavioral Health Care, an assistant professor in the Department of Medical Ethics and Health Policy, Perelman School of Medicine, and a senior fellow of the Leonard Davis Insititute of Health Economics, University of Pennsylvania.