The federal government can — and should — help the mentally ill

Last week, the Congressional Budget Office (CBO) released a series of cost estimates for repealing an archaic federal policy known as institutional exclusion for mental illnesses (IMD). The rule has led to a shortage of mental hospital beds across the country, associated with hundreds of thousands of mentally ill people currently in prison or on the streets. Even the highest cost estimated by the CBO for the repeal of the IMD exclusion is justifiable. The federal government should finally put an end to this antiquated policy.

Passed as part of Medicaid in 1965, the IMD exclusion largely prohibits federal Medicaid dollars from funding treatment provided by IMDs — IMDs generally referring to psychiatric hospitals with more than 16 beds. The rule was intended to prevent states from shifting the financial burden of their mental health systems — which were then asylums — to the federal government. While states must foot the bill for IMD-based care, their costs are shared with the federal government for services provided elsewhere.

IMD exclusion catalyzed the deinstitutionalization of psychiatric patients from hospitals to the community. Still in place, it continues to create a financial disincentive for state investment in bed capacity, preventing a full continuum of mental health care.

The consequences of too few hospital beds are obvious. Psychiatric patients who need intensive and acute medical services are instead confined to emergency departments or go through several hospital discharges without stabilization. Estimates suggest nearly 218,000 homeless people are either seriously mentally ill or suffer from chronic drug addiction, and those in jails and prisons are three to five times more likely than the general population to have the disease serious mental. All this despite an almost 1,500% increase in spending on community mental health services between 1981 and 2015.

A Kaiser Family Foundation senior policy analyst correctly diagnosed the circumstances in a commentary: “There has been an understanding over the past few years that this lack of federal funding contributes to high levels of unmet need. Unmet need is not surprising given a 95% drop in bed capacity — which continues to decline — since peak levels around 1955. Recognizing the ramifications, lawmakers on both sides have shaken the rule through of various exceptions since its inception.

The CBO’s new cost analysis comes ahead of the September expiration for one such exception: Thanks to the SUPPORT Act passed in 2018, states can modify their Medicaid plans and receive federal matching funds for enrollees with substance use covered in IMDs; the impetus being to expand access to treatment for opioid addiction. The CBO presents federal cost estimates for two main policy options: permanently extend this expiring “state plan option” and repeal the IMD exclusion. Cost estimates for variations of each policy are also presented.

The most expensive option, for the complete repeal of the IMD exclusion, may initially seem costly: $38.4 billion over 10 years. But the analytical approach providing this estimate makes this first impression misleading for three main reasons.

First, the CBO acknowledges that under current law, states may seek exemptions to the IMD exclusion through various channels, such as federal waivers. As more states enact exemptions, the increased associated federal expenditures would account for a portion of the dollars in the repeal cost estimate. In other words, the federal government will end up paying part of this amount even if it does nothing and the current law remains as it is.

What part of the cost will be paid anyway? The CBO does not elaborate, saying it depends in part on effects that will differ between states with different laws, and possible adjustments to the use of mental health services. However, the CBO expects that by 2033, 80% of Medicaid enrollees would live in states with IMD exclusion waivers that make treatment for serious substance use disorders and mental illnesses eligible for matching federal dollars when provided in IMDs.

Second, while the estimate incorporates cost offsets from projected decreases in emergency department use (a welcome addition not included in a previous estimate), the CBO does not comprehensively incorporate cost savings that will no doubt be realized in other systems strained by continued deinstitutionalization, such as criminal justice and homeless services.

In a striking example, as documented by my colleague Stephen Eide for New York, between fiscal years 2014 to 2018, New York City added six new homeless shelters – from a base of 22 – specifically dedicated to shelters for the Mental Health. The total number of mental health shelter beds in the city is greater than the combined total of adult beds in New York State’s public psychiatric hospitals and psychiatric beds in NYC Health + Hospitals facilities.

Finally, the CBO’s estimate is based in part on analysis of the effects of current state waivers, finding that federal spending for Medicaid enrollees has increased in states with waivers. Indeed, more IMD stays were reimbursed (as expected), more providers accepted Medicaid payment, and more of these services were used. In addition, emergency room visits have been reduced. In other words, exactly what we could hope to happen happened: unmet needs were met and unnecessary emergency room hospitalizations decreased. These are spending increases that are expected, wanted and beneficial. And in total, the total estimated increase in repeal spending would be less than 0.5% of projected federal spending on Medicaid in 2033.

Given the above, full repeal is the most sensible of the policy options proposed by the CBO. It is true that exemptions to the IMD exclusion are currently available to states that request them and meet their requirements. In some ways, these requirements can serve as a mechanism to avoid unnecessary expense. But states have limited resources to adopt new programs and systems; the availability of such exemptions, on a large scale, also did not seem to have remedied the inadequacy of care and access to it.

The CBO includes separate estimates for repeal variants that exempt people with substance use disorders or serious mental illnesses from the IMD exclusion. In short, exempting one group and not the other, on any given basis or justification other than cost, would be discriminatory. The IMD exclusion itself is a discriminatory policy – against those for whom treatment is most appropriate provided in a hospital setting. Reimbursement for registrant health care services should be based on medical necessity, not diagnosis or location.

Although the CBO’s estimates are conservative and likely exclude many cost offsets, an effective mental health system that provides seriously mentally ill people with high-quality services – both inpatient and outpatient – ​​will not come for free. The associated cost benefits, however, are realized both by the mentally ill and by society as a whole, through safer streets and cities and more efficient public systems and services.

Carolyn D. Gorman is a Research Associate at the Manhattan Institute, where her research examines the impact of policy changes in the US healthcare and justice systems on people with serious mental illness.