Medicaid’s Institutions for Mental Diseases (IMD) Exclusion Rule: A Policy Debate—Argument to Repeal the IMD Rule

Transforming mental health care will require dramatic policy changes—akin to civil rights legislation of the 1960s—to fully integrate mental and physical health care. These changes will require full and equal consideration of both mental and physical health and disease—a shift that requires far more than insurance coverage parity. What is required is a paradigmatic shift in how we conceive of mental health and illness to recognize the fact that mental illnesses exist within the same ontological realm as physical illness. Mental illness is illness and there is no health without mental health. This is what we call “conceptual parity” (1).

Repeal of the IMD rule would serve to advance conceptual parity by allowing for the development of appropriate clinical settings for individuals with serious mental illness. In the limited cases in which federal funding has flowed to IMDs, evidence suggests that the change in policy resulted in improved quality of care. Provisions in the Affordable Care Act (ACA) required the Centers for Medicare and Medicaid Services (CMS) to conduct a demonstration project to assess the effects of providing Medicaid reimbursements to private psychiatric hospitals, which are considered IMDs under the current rule. The Medicaid Emergency Psychiatric Services Demonstration Evaluation (MEPD) ran from 2012 through 2015. Eleven states and the District of Columbia participated. Cost data limitations prevented CMS from expanding the program, but the final report yielded some limited insights regarding the use of IMDs (2).

Concerns over warehousing of patients appear to be unfounded. Across more than 16,000 admissions to IMDs, the median length of stay was seven days, 89% of stays lasted fewer than 31.4 days, and the vast majority of patients were discharged to their homes rather than to other facilities (2). Furthermore, interviews with Medicaid beneficiaries and program staff found that MEPD may have improved the quality of care: enrollees overwhelmingly reported satisfaction with the quality of care they received at IMDs, and state and facility staff believed that the demonstration improved access to higher quality psychiatric care (2). The temporary waiving of the IMD exclusion of private psychiatric hospitals did not result in inappropriate institutionalization or decreased funding for community-based interventions. It simply expanded the supply of inpatient beds available to Medicaid beneficiaries and removed barriers to care. Therefore, the tendency to frame investments in inpatient care and community care as a zero-sum scenario appears to be misguided.

In addition, some states have applied for and received Section 1115 innovation waivers to treat Medicaid beneficiaries in IMDs (Table 1). Requests to waive the exclusion for substance use disorders and mental disorders are among the most common type of waiver applications. Seven states have received approval to receive Medicaid reimbursement for substance abuse services in IMDs, and one state has authority to receive Medicaid reimbursement for mental health services in IMDs. Five additional states are seeking authorization to provide substance abuse treatment for Medicaid beneficiaries in IMDs, and two have pending applications to provide mental health care in IMDs. Applications to waive the IMD exclusion rule come from politically and geographically diverse states, ranging from Arizona and West Virginia to California and Vermont (3).

These waivers carry the stipulation that IMDs cannot supplant community-based services, and applications to waive the IMD rule are frequently paired with waivers to provide additional community-based care. Given the current structure of Medicaid as state administered and jointly funded, waiving the IMD rule simply gives state health administrators another tool for treating a subset of acute patients.

Repealing the IMD rule is fiscally responsible, medically appropriate, and ethically defensible. When hospitalization is required, the current Medicaid reimbursement structure incentivizes the use of psychiatric facilities with fewer than 16 beds and inpatient care in nonspecialized units throughout hospitals. Financially, limiting care to small facilities is less efficient. Because certain administrative costs remain fixed as facility size increases, the IMD rule prevents public payers from taking advantage of economies of scale. Furthermore, providing care for psychiatric emergencies in nonspecialized hospitals fails the basic ethical obligation to provide individuals with care in the most appropriate setting. Persons with psychiatric emergencies ought to be treated in psychiatric settings staffed by well-trained behavioral health professionals.

Proponents of the IMD rule argue that it is essential to prevent states from shirking their responsibilities to persons with mental illness through over-reliance on long-term, inappropriate institutionalization rather than investment in community-based services. Yet, warehousing of individuals with mental illness continues unabated through mass incarceration. The opioid crisis has increased the need for structured care settings, and the advent of more effective modalities of psychiatric treatment has reduced inappropriate institutionalization. Ultimately, the IMD rule is a policy mismatched in both time and place. It violates conceptual parity without encouraging adequate care across the mental health continuum, either in communities or in inpatient settings.

Many critics of the IMD rule cite the rule as a primary cause of inpatient psychiatric bed shortages, whereas those who continue to defend the rule argue that improved medication-assisted treatment, decreased lengths of stay, and treatment in outpatient and community settings can provide adequate mental health treatment and reduce the need for inpatient beds. The reality of treatment capacity in the United States is far more nuanced and likely a blend of the two views. However, on straightforward observation we may note that even with sufficient funds for a comprehensive community psychiatry system, there will remain a proportion of seriously ill people who require structured inpatient care settings.

The extent and distribution of psychiatric bed shortages remains a subject of debate. The past half century has been marked by deinstitutionalization of persons with mental illness. In 1955, there were nearly 560,000 state hospital psychiatric beds; today there are just over 37,000 public inpatient beds and nearly 36,000 private beds. Much of the downsizing is attributable to appropriate shifts in care modalities, such as medication-assisted treatment and decreased reliance on prolonged institutionalization. Other exogenous factors, such as the adoption of managed care in the 1990s and lower profit margins for hospital psychiatric units compared with medical-surgical wings, have also contributed to downsizing psychiatric wards. Fewer inpatient beds on a national per capita basis does not necessarily mean there is a national shortage of beds. Some communities may have adequately calibrated capacity, and others may have a surplus (4).

However, the pendulum may have swung too far in some localities. In 2015, a survey by the National Association of State Mental Health Program Directors found that 35 of 46 states experienced shortages of psychiatric hospital beds (5). Of the 35 states reporting bed shortages, 25 reported increased waiting lists for state hospital beds, and 16 reported increased wait times for beds in private psychiatric institutions and general hospital psychiatric units. Given the federal policies that disincentivize institutional care, state governments typically respond to inpatient bed shortages with community-based alternatives, such as funding residential crisis beds and increasing assertive community treatment. Such workarounds may provide adequate care in some cases and reduce inpatient hospitalizations, but they fail to address the significant problem of lack of inpatient beds for those who need them.

The litany of complaints about gaps in care for persons with mental illnesses does not begin and end with the IMD rule, but several care shortcomings related to capacity constraints are linked to the rule—in particular, over-reliance on less appropriate care in emergency departments (EDs) and unspecialized beds in general hospitals. Recent evidence indicates that too much of America’s mental health care occurs in EDs and nonspecialized beds distributed across hospitals (known as “scatter beds”), rather than in psychiatric units of hospitals or in IMDs. Research conducted for the Substance Abuse and Mental Health Services Administration estimated that scatter beds account for 36% of general hospital mental health expenditures (6), and an American College of Emergency Physicians (ACEP) survey of ED medical directors found that 81% believed that dedicated emergency psychiatric facilities would improve care (7). For persons whose care requires a hospital admission, scatter beds cannot provide the specialized treatment required—and provided—by psychiatric hospitals.

Furthermore, reliance on EDs for mental health crises has increased. A 2016 study by the Agency for Healthcare Research and Quality found that from 2006 to 2013, ED visits per capita increased 52% for psychoses or bipolar disorder and 37% for substance use disorders (8). Many of these visits are preventable. In 2014, 30-day all-cause readmission rates after hospitalization for schizophrenia and other psychiatric disorders were 15.8% among the privately insured but 24.9% among those with Medicaid (9). Medicaid beneficiaries have higher readmissions rates than individuals with private insurance across all hospitalizations, but the gap is nearly twice as wide for mental health admissions (10).

High readmission rates reflect the lack of appropriate care in EDs. In EDs, psychiatric patients are more likely to wait for extended periods (ranging from a single day to weeks) in beds, hallways, and locked rooms until inpatient beds become available—a practice euphemistically referred to as “boarding” (11). For psychiatric patients in particular, stress-inducing and restrictive EDs can cause deterioration, and an ACEP survey found that 62% of EDs provided no psychiatric services to patients who were boarded (7).

Although there are no systemic data on the prevalence of boarding, recent studies suggest a widespread and growing problem that causes disruption in care. ACEP found that 80% of ED medical directors reported boarding psychiatric patients, and 90% of medical directors boarded psychiatric patients at least once per week, and over half boarded patients on a daily basis (7). The U.S. Department of Health and Human Services, ACEP, and the Joint Commission have all expressed concern over the prevalence and effect of psychiatric boarding and have identified insufficient inpatient bed supply as a common cause of the practice (12–14). Addressing boarding, scatter beds, and other disjunctions in the mental health care system will require significant shifts in care delivery, and the IMD rule’s curtailment of financing options for inpatient care is one of several barriers to essential reform.

Providers, payers, policymakers, and the public would balk at arbitrary care restrictions for any other illness. There are no comparable limitations on institution size for inpatient oncologic or cardiac care. And although mental health parity typically refers to uniformity of insurance coverage, conceptual parity between mental illness and physical illness means that mental illness is recognized as illness. Conceptual parity should be the ultimate goal (1).

The road to conceptual parity requires eliminating the IMD rule. Meeting our nation’s enormous mental health care needs requires a comprehensive continuum of services, unencumbered by ideological commitments and facile appeals to the worst aspects of psychiatry’s history. There is now compelling evidence that allowing the federal government to pay for Medicaid beneficiaries to receive treatment in IMDs improves care—without resulting in the deplorable conditions of the past (15).