Busting the Deinstitutionalization Myth: We Actually Have More Beds Than Ever Before

Many people who know almost nothing else about the mental health system can nevertheless recount the story of the “failure” of “deinstitutionalization” in America. The story is repeated so often that it’s widely accepted as if it were a famously indisputable math formula:

Large state hospital asylums started closing decades ago, but promised community beds and services never came. As a result, today, there’s a disastrous bed shortage and huge populations of untreated, severely mentally ill people are homeless or in prisons.

Any numbers provided are close to these:

In the 1950s, there were about 550,000 state hospital psychiatric beds, or 330 beds per 100,000 people. Today, there are only 37,000 state hospital psychiatric beds, or about 11 beds per 100,000 people.

For the past two decades, this story has been regularly re-told everywhere from popular right-wing periodicals like The Wall Street Journal, National Review, and Breitbart, through mainstream and left-liberal sources like The New York Times, NPR, PBS, The Daily Beast and The New Yorker, to investigative news outlets like Mother Jones and Kaiser Health News. The story also plays a central role in political lobbying by mental health organizations and providers, in Democrat and Republican platforms, and in public education by the Substance Abuse and Mental Health Services Administration (SAMHSA).

Essentially, this deinstitutionalization disaster story has become a culture-wide dominant narrative. It has fueled beliefs that compassionate improvement of the mental health system—and help for the unhoused and imprisoned—requires bringing back more psychiatric beds and coercively treating more people. Even those who critique pro-force sentiments in outlets like The Nation nevertheless frequently echo the dominant narrative’s basic elements.

However, in 2017, the U.S. National Association of State Mental Health Program Directors (NASMHPD)—the people who actually oversee America’s mental health systems—quietly fact-checked this dominant narrative. Their resulting report showed that these oft-repeated bed numbers are not only inaccurate but wildly misleading.

And from the real numbers, there emerges a fundamentally different portrait of America’s mental health system, its impacts on society, and what’s gone wrong. It’s a picture of an America where there’s never in history been more psychiatric beds per capita, or more widespread psychiatric monitoring and coercion.

Where and why did the viral disinformation start?

If you’ve ever been a psychiatric patient or worked in community mental health in America or, like me, regularly talked with people from both these groups, something quickly becomes obvious: State hospitals rarely get mentioned. Much more commonly, people mention general hospital psychiatric wards, private psychiatric hospitals, group homes, assisted living facilities, nursing homes, and other institutions and services. And in fact, only 1.6 percent of people getting public mental health care in 2020 got it in state hospitals.

So, why has so much public attention focused on state hospital bed numbers?

Reading the above-cited news stories, the answer becomes obvious: Every one cited, quoted, was co-written, or otherwise involved the Treatment Advocacy Center (TAC), its oft-quoted report about declining state hospital bed numbers “Going, Going, Gone,” and/or TAC’s founder, psychiatrist E. Fuller Torrey. And not coincidentally, rivaled perhaps only by the National Alliance on Mental Illness, TAC and Torrey have long been this country’s most well-funded and influential lobbyists for expanding forced psychiatric treatment. Indeed, when New York Mayor Eric Adams recently called for more aggressive uses of psychiatric detention powers, The New York Times dubbed it a “coronation” for TAC and Torrey.

TAC and Torrey have focused attention on eye-catching state hospital bed-number declines to help push their campaigns for expanding coercive treatment, and they’ve been extremely successful. In my book Your Consent Is Not Required: The Rise in Psychiatric Detentions, Forced Treatment, and Abusive Guardianships, I show the consequences: Rates of coercive psychiatric interventions have been rising dramatically for several decades as broadening mental health laws get used against ever-wider ranges of people and in many institutions from schools, government housing, and workplaces to mental health hotlines and nursing homes.

So, how can coercive psychiatric interventions be rising amid a purported catastrophic shortage of beds? The answer is that “bed capacity” in the United States has steadily expanded in step with that rise.

Resurgent interest in the facts

TAC and Torrey’s story has rallied support for increased mental health funding and more aggressive forced drugging laws, but it’s always been too far divorced from reality to guide on-the-ground, day-to-day decision-making. Knowledgeable mental health system leaders know that most beds and services today are outside state hospitals, and that expanding hospital psychiatric care is, by far, the most expensive of all possible options. And even the American Psychiatric Association (APA) quietly recognizes that scientific evidence of the helpfulness of psychiatric hospitalization is “absent.”

So, no new large asylums have been built. Instead, under mounting public pressure to solve the nation’s seeming “mental health crisis,” mainstream mental health system leaders are showing a resurgent interest in understanding how many psychiatric beds of which types actually do exist in America, and may optimally be needed.

One question that needs answering first is, “A half-century after most asylums closed, what is a psychiatric bed today?”

A team of researchers from the RAND Corporation recently published a viewpoint in JAMA Psychiatry, “Estimating Psychiatric Bed Shortages in the US.” The authors described it as “worrisome” that “there are no standardized approaches or best practices” for determining what a “psychiatric bed” is—let alone how many are needed or what causes an apparent “shortage.”

Broadly, publicly funded psychiatric beds can be divided into two main categories:

Category 1: Inpatient beds in hospitals or hospital-like settings that are able to provide 24/7 intensive psychiatric care.

Category 2: Residential beds in the community where people labeled with mental illnesses live in home-like settings and receive any of varying amounts of publicly funded voluntary or involuntary mental health treatments and practical supports.

In the 1950s, asylums were the predominant location for both of these categories of beds. Today, there are many variants of both, and the divisions are not always clear cut—for example, residential treatment centers are inpatient institutions that also have residential beds.

There also aren’t standardized, completely reliable ways to find and count psychiatric beds, the RAND authors explained.

Generally, a facility’s bed numbers are included in licensing information—however, at any time that number may not correspond to staffed beds, different types of facilities may be overseen by different government departments, and some common bed types, such as supportive housing for people labeled with mental illnesses, may not require licensing at all. One can also count psychiatric “patients in beds” on one selected day—but any day may not accurately reflect other days or full capacity. Another method is to count an institution’s “discharges” linked to treatment for mental disorders.

Then, to understand shortages, one has to track how different types of beds and services interact. For example, there may appear to be a shortage of hospital psychiatric beds when there’s actually a shortage of supportive housing to move people into when they’re ready for discharge from hospitals. Or, when patient case managers don’t have access to frequently updated directories of all available beds in their area, then shortages can seem common even when they’re not.

Consequently, counting psychiatric beds and illustrating trends across time and population changes involves “cobbling together” data from different sources, drawing estimates from studies and surveys that may have been done in different years in different ways, and “roughing in” calculations. Yet, the very complexity highlights the operational importance of gaining better understanding.

So, the NASMHPD decided to take on the task of counting psychiatric beds in America. And their data and analyses make it clear that the dominant narrative about deinstitutionalization is completely off-base.

The reports that are shifting the dominant narrative

As of 2023, there are four related reports that, together, have started to shift the dominant narrative.

Most importantly, published in 2017 by NRI, the research institute of the NASMHPD, the report “Trends in Psychiatric Inpatient Capacity, United States and Each State, 1970 to 2014” provided a comprehensive, nation-wide count of beds in category 1—inpatient psychiatric beds in hospitals or hospital-like settings.

In 2019, SAMHSA issued “Civil Commitment and the Mental Health Care Continuum.” This report drew on the NASMHPD report’s data and examined how psychiatric bed numbers and types have changed historically alongside changing standards of coercive care.

In 2022, the APA released “The Psychiatric Bed Crisis in the U.S.: Understanding the Problem and Moving Toward Solutions.” The APA also drew on the NASMHPD report’s data and began to do modeling of how many beds of which types might be required to provide an optimal continuum of care in a typical community.

Later in 2022, the NASMHPD published an update to its own report, showing bed-number trends to 2018.

In this review and analysis, I’ll refer to these reports as the NASMHPD, SAMHSA, and APA reports. I also interviewed Ted Lutterman, lead researcher for the NASMHPD reports, in 2021 for my book, and again in 2023 for this article.

In conversation, Lutterman has repeatedly made the gist of his findings clear: “E. Fuller Torrey or others are always talking about how there used to be half a million people in state hospitals, and if we had that number of beds today, at equivalent rates, we’d need so many beds.” However, Lutterman said, the widely cited numbers promoted by Torrey, TAC, and many politicians and media outlets “just don’t add up.”

Both the APA and SAMHSA reports also challenged TAC and Torrey’s main claims. The SAMHSA contracted-researchers accused “E. Fuller Torrey and colleagues from the Treatment Advocacy Center” of promoting bed numbers and comparisons that were neither “apt” nor truly “meaningful.” Irked by inflammatory claims that prisons are “the new asylums,” the researchers countered that “nearly 20 times more people with a serious mental illness today receive care in the nation’s public mental health systems than are housed in its jails and prisons.” In a similar vein, the APA report noted that every study that has actually tracked discharged asylum patients found that only miniscule percentages ever became homeless or imprisoned.

So, what do the real numbers show?

First, there were nowhere near as many psychiatric beds in the 1950s as Torrey and TAC have continually claimed.

How many inpatient beds existed in the era of large asylums

In the 1950s, nearly all psychiatric beds in America were in state/county hospitals. The NASMHPD reports pointed out, however, that state hospitals in that era held many non-psychiatric patients as well. The SAMHSA report similarly acknowledged (as did the APA report) that “large swaths” of state hospital patients used to include people with tuberculosis, epilepsy, syphilis, elderly dementia, alcohol problems, and intellectual or developmental disabilities. These are people who today receive services in other types of facilities that have themselves increased immensely in number over the decades, such as ordinary medical hospitals, nursing homes, “sober homes,” and institutional homes for people with intellectual and developmental disabilities.

Focusing on psychiatric beds only, Lutterman found reliable data showing that, even as late as 1969, only 58 percent of state hospital patients were psychiatric patients. So, the actual number of state hospital psychiatric beds in 1969 was 42 percent lower than the number commonly cited in the dominant narrative.

Although the percentages from the 1950s appear to roughly parallel the 1969 numbers, Lutterman told me that he was reluctant to use that data due to the murkiness of both psychiatric diagnoses and state hospital admission criteria in that era. I asked him if he generally believed the true percentage of non-psychiatric beds in state hospitals would have been smaller or larger in the 1950s. “It’s unclear, but it sounds like it would be even larger,” Lutterman answered. “[State hospitals] were sort of a catch-all place [for people] that society didn’t know what else to do with.”

So, there were not 550,000 state hospital psychiatric beds in the 1950s, but more likely, at most, 58 percent of that number—about 319,000 state hospital psychiatric beds. Roughly accounting for national population changes during that decade, this was about 190 state hospital psychiatric beds per 100,000 people in the 1950s.

And by comparison, how many inpatient psychiatric beds exist in America today?

Counting category 1: The number of inpatient psychiatric beds that exist today

As the so-called “deinstitutionalization” movement gained traction from the 1960s to ‘80s, large state and county hospitals were closing. However, contrary to the dominant narrative, smaller psychiatric hospitals, institutions, and community-based beds, facilities, and services began multiplying and receiving enormous increases in funding.

In reviewing this history in the NASMHPD reports, Lutterman focused on counting “inpatient” beds—hospitals or hospital-like residential facilities. Aside from state/county hospitals, these include:

• Private psychiatric hospitals
• General hospitals with psychiatric wards
• Veterans Administration (VA) medical centers
• Community mental health centers
• Residential treatment centers
• Other crisis care facilities

VA and state hospital bed numbers have indeed declined a lot, but Lutterman found that bed numbers (or the total number of patients on a selected day) in many other facilities increased considerably from 1970 to 2018. Beds in private psychiatric hospitals increased nearly five-fold to 54,396. Residential treatment center beds nearly tripled to 36,845 beds. Beds in general hospitals more than doubled to 40,530.

Among these, Lutterman found 187,877 inpatient psychiatric beds in America in 2018. That’s about 57 beds per 100,000 people—five times as many inpatient psychiatric beds as the “11 beds per 100,000” number that is commonly publicly cited in deinstitutionalization disaster stories.

And Lutterman found still more.

Nursing homes exploded in number in the late 1950s and ‘60s, supported by enabling legislation and the launch of Medicare and Medicaid. Although mainly for frail seniors, the reports from NASMHPD, SAMHSA, and the APA all acknowledged that nursing homes have become a common inpatient setting where psychiatric patients also get placed. These include large numbers of adults who are not elderly and/or do not have dementia or disabling physical conditions. No one is formally tracking how many ordinary psychiatric patients stay in nursing homes, but in some states thousands have been identified in Department of Justice lawsuits targeting them as unnecessarily restrictive, coercive settings for most psychiatric patients.

Since these are usually people diagnosed with serious mental illnesses, Lutterman used data quantifying the number of nursing home residents diagnosed with what are typically regarded as serious mental illnesses: bipolar disorder or schizophrenia. These comprised 13 percent of nursing home residents. Lutterman advised “caution” about the precision of the number, but this calculation added 223,917 patients—another 67 beds per 100,000 people.

The table below shows the main types of inpatient psychiatric beds in America that Lutterman could reliably quantify—with corresponding calculations of beds per 100,000 people to facilitate comparisons over time relative to population growth.

Number and Rate per 100,000 population of psychiatric inpatients and other 24-hour residential treatment patients at end of year 2018. (Excerpted with decimal-rounding adjusted from Table 1 and Table 5, “Trends in Psychiatric Inpatient Capacity: United States and Each State, 1970 to 2018.”)
Type of Organization	Patients in Inpatient and Other 24-Hour Residential Treatment Beds at End of Year
	Residents/Beds	Rate Per 100,000 Population
State & County Psychiatric Hospitals	35,725	10.9
Private Psychiatric Hospitals	54,396	16.6
General Hospital with Separate Psychiatric Units	40,530	12.4
Veterans Administration Medical Centers	6,992	2.1
Residential Treatment Centers	36,845	11.3
Other Specialty Mental Health Providers with Inpatient/Residential Beds	13,389	4.1
Nursing Homes – patients with diagnosis of schizophrenia or bipolar disorders (2019)	223,917	67.1
TOTAL	411,794	124.5

 

In total, the NASMHPD report identified 411,794 inpatient hospital and hospital-like psychiatric beds in America, or 124.5 beds per 100,000 people.

This is now more than ten times as many inpatient psychiatric beds as the “11 beds per 100,000” number that has been most commonly publicly cited.

By direct comparison, then, in 2018, America had more inpatient psychiatric beds than in the 1950s—411,794 beds compared to about 319,000. Accounting for population growth, in the 1950s, there were about 190 state hospital psychiatric beds per 100,000 people while, in 2018, there were 124.5 beds per 100,000 people.

So, there’s been at most a one-third reduction in inpatient bed numbers per capita between the 1950s and 2018. It’s not insignificant, but hardly the catastrophic 96 percent decline that deinstitutionalization disaster stories claim.

This is where Lutterman stopped formally counting. However, he pointed to many more types of psychiatric beds where the numbers were difficult to quantify, but likely very substantial. One of these was category 2, community residential psychiatric beds—and wherever we can find numbers for them, they are revealing.

What community residential psychiatric beds are and why they’re important

Providing long-term residency, with ongoing psychiatric care and support with daily living, was a key role that state hospitals played for many people in the 1950s—and, as discussed extensively in all four reports, moving those patients into community-based residences was a primary goal of deinstitutionalization.

Today, such residences can include, for example, publicly funded private or state-owned small group homes, large congregate care housing, independent-living or assisted-living facilities, supportive housing, government-subsidized apartments, and so on. When these are tailored and funded as psychiatric beds, residents will be getting anywhere from low- to high-intensity monitoring, mental health interventions and treatments, and practical daily supports—often via “Assertive Community Treatment” (ACT) or similar “case management” teams.

Like psychiatric hospitals in the 1950s and today, community residential psychiatric beds can vary in quality—but it’s not possible to reasonably discuss deinstitutionalization without including them.

I asked Lutterman if he agreed it was reasonable to say that, to get an accurate total count of psychiatric beds and to compare today with the 1950s, community residential psychiatric beds must be included. “It is,” Lutterman answered. “The question is, how to get that number?”

Community residential psychiatric beds are not nationally tracked. And depending on the type of facility or residence, Lutterman explained, it may not be licensed by any government agency at all. It may be owned or rented by the client/patient or a private company rather than a public agency, and the housing subsidies, psychiatric services, and supports may be coming through any of a variety of funding streams at local, county, state or federal levels.

Consequently, a reasonably functional “continuum of care” and plan for addressing bed shortages in any community would require having a centralized, frequently updated listing or data repository of currently available beds of all types—yet stunningly, few states have them. In 2019, SAMHSA and the NASMHPD launched a collaboration to help states begin building electronic registries of psychiatric beds. These efforts are still nascent in most states, though.

I found five states that were able to provide what appeared to be reasonably organized and reliable data on community residential psychiatric beds.

Counting category 2: Numbers of community residential psychiatric beds

Connecticut has a bed registry dashboard and New Jersey provided data showing, respectively, about 29 and 39 adult community residential psychiatric beds per 100,000 people in those states. Maryland provided a list that identified 46 adult beds per 100,000 population. However, it became apparent that these were likely low-end, partial numbers, when I found states with more comprehensive tracking of other common types of beds as well—notably, these are the two states where the public clamor about alleged bed shortages has been the loudest in recent years.

The New York State Office of Mental Health (OMH) data dashboard describes and tracks many types of specialized, supportive community housing for people labeled with serious mental illnesses, including “Apartment Treatment,” “State-Operated Community Residence,” “Supported Single Room Occupancy,” and more. According to an OMH spokesperson, and confirmed in the dashboard, the state “funds more than 48,000 units of housing for individuals with mental illness.” In New York, these add 246 adult community residential psychiatric beds per 100,000 people.

In California, the state’s community bed tracking is poor, but it does carefully track its primary community-based psychiatric support system, called “Full Service Partnerships” (FSPs). As of 2021, there were 71,384 people enrolled in FSPs. These programs provide housing as needed and “intensive” psychiatric treatments and other supports for people who are labeled as having “severe mental illness” and at risk of homelessness. According to UCLA’s Sam Tsemberis, who has studied FSPs, these people are typically getting housed in unlicensed, subsidized locations like apartments, rooming houses, and independent living facilities—but, he estimated, as many as 15-30 percent of clients remain essentially unhoused. So FSPs represent closer to about 55,000 beds or 141 community residential psychiatric beds per 100,000 people in California.

About one-third of California’s FSP clients are children and youth, of which many are in foster family homes—a reminder that all states also have community-based residential psychiatric beds in their foster-care systems. A recent study found that, in 2016, some 35 percent of children in foster care were also getting psychiatric care. That same year, about 374,000 children and youth in total were living in group or family-style foster homes (as distinct from hospital-like foster-care institutions). This would then comprise about 131,000 community residential psychiatric beds, or 40 beds per 100,000 people.

In all five of these states, the psychiatric bed count now surpasses the average number of beds per 100,000 people in the 1950s. Before we do a complete tally, though, it’s important to briefly mention some of the other beds that are still missing from the count.

Large numbers of other inpatient and residential psychiatric beds

There were other major types of beds that were not included in Lutterman’s counts because he couldn’t find sufficiently clear numbers for them. In some cases, there were also questions as to whether the beds should “appropriately” be counted as psychiatric beds. Here, the lack of standardized definitions for what constitutes a “psychiatric bed” becomes relevant. For example, many people today get involuntarily psychiatrically detained in hospital emergency room beds for significant periods. Yet some might argue that emergency room beds shouldn’t be counted, because chaotic emergency departments are inappropriate settings for psychiatric care. But by that logic and standard, we’d have to exclude any “inappropriate” facilities from these counts—starting with the many reportedly decrepit, overcrowded, chaotic, abusive state hospitals of the 1950s.

With that in consideration, these are some of the other major types of public beds earmarked today for people labeled with serious mental illnesses that have not yet been counted (see the Appendix for more details about these types of beds and where the estimates come from):

• Many beds for children and youth in inpatient-like residential psychiatric facilities (as distinct from family-style foster homes) exist outside common tracking systems—including in the so-called “troubled teen industry” and out-of-state institutional foster-care settings. This could be another 158,000 inpatient-residential psychiatric beds.
• In prisons and jails, there are both formal inpatient psychiatric units and the equivalent of residential psychiatric beds—together, these appear to comprise about 288,000 beds.
• “Scatter beds” are general beds in community hospitals that sometimes get used for psychiatric care. These beds hosted about 121,000 patients in 2014.
• Many mental health visits to hospitals, voluntary and involuntary, result in people getting “boarded” in emergency room beds. Boarding is usually defined as periods lasting from six hours to days, weeks, or months—already in 2008, this comprised 1.7 million patients annually.
• An unknown percentage of some 988,000 people in nursing homes diagnosed with either depression or anxiety disorders or both may actually be ordinary psychiatric patients (without dementia or debilitating physical conditions). Even a low percentage could represent a significant number of beds.

Getting to the totals: Adding inpatient and community residential psychiatric beds together

Based on the national averages for inpatient beds from the NASMHPD reports, and the other data gathered, we can now do a final tally (see the table below).

In America’s state hospital asylums in the 1950s, there were about 190 psychiatric beds per 100,000 people. Today, there appear to be between 193 and 410 psychiatric beds per 100,000 people. That’s anywhere from slightly more beds to more than two times as many beds per capita as existed in the 1950s. Adding in the best estimates for missing child and youth beds and prison beds, the number today is between 328 and 545 psychiatric beds per 100,000 people—approaching three times as many psychiatric beds as existed in the peak era of state asylums.

If we had reliable ways to calculate for the number of scatter beds, missing nursing home beds, and 1.7 million or more psychiatric patients in emergency rooms each year, the total might skyrocket beyond what existed in the 1950s.

National-Level Data (from various years)
Type of facility or bed	Source of data (see Appendix for more details)	Known or estimated number of beds	Beds per 100,000 population
Inpatient psychiatric beds	(NASMHPD Report up to 2018)	411,794	124
Community residential psychiatric beds for children and youth in family foster-care	(HHS AFCARS and Keefe study of 2016)	131,000	40
Community residential psychiatric beds	(state-level bed numbers in 2023)	(see state-level bed numbers table)	29 – 246
Subtotal			193 – 410
Additional inpatient-residential psychiatric beds for children and youth	(est. 158,000 long-term patients annually in 2008)	158,000	48
Prison and jail inpatient psychiatric unit beds and residential psychiatric beds	(32,000 California patient-inmates in 2016 extrapolated nationally)	288,000	87
Inpatient scatter beds in general hospitals	(est. 121,000 patients annually in 2014)	?	?
Nursing home beds for people labeled with depression and/or anxiety disorders	(unknown percentage of 988,000 beds in 2018)	?	?
Emergency room beds	(1,700,000 visits/patients annually in 2008)	?	?
TOTAL			> 328 – 545

 

 

State-Level Data (2023)
Public community residential psychiatric beds by state	Number of beds	Beds per 100,000 population
Connecticut	1,025	29
New Jersey	3,546	39
Maryland	2,829	46
New York	48,000	246
California	55,000 (minus unhoused, includes children)	141

 

There are still many estimates and gaps in these numbers. Yet, at the very least, it is abundantly clear that the dominant narrative is completely misleading.

So, what does it all mean?

America’s expanding systems of psychiatric coercion

The dominant narrative has long stated that America has 96 percent fewer psychiatric beds than in the 1950s which, combined with a desperate lack of funding for community beds and services, has led to legions of untreated, seriously mentally ill people “falling through the cracks” of the system and ending up homeless or in prisons.

This deinstitutionalization disaster story has been extremely effective for generating public support for increased mental health funding and expanded coercive interventions, while excusing the system’s every visible, chronic failing. It’s also given conservative and liberal politicians and journalists alike a simplistic, convenient narrative that can, at will, divert criticism away from innumerable other issues and policies that are worsening inequities, injustices, and violence across society.

This dominant narrative is so popular, even the mainstream organizations that have helped debunk it seem reluctant to let it go completely. Neither the NASMHPD, SAMHSA, nor the APA have been vigorously promoting the true bed numbers; worse, to varying degrees, their reports all partially disguised them. For example, the APA report acknowledged that state hospitals historically served large populations of non-psychiatric patients—and then completely ignored that fact in its own psychiatric bed-number calculations and comparisons. The true state hospital bed numbers are clearly presented in Lutterman’s calculations and discussions—but appear in none of his more visually prominent bed-number comparison tables. Though all four reports emphasized the immense growth and central importance of community residential psychiatric beds in the continuum of care, none of the reports included any attempts to estimate and add in their numbers. Treatment Advocacy Center itself was a co-author and centrally involved in a related 2017 NASMHPD report that drew on Lutterman’s numbers; nevertheless, TAC has not revised its own issue briefings or reports about bed declines that include only the misleading numbers about state hospitals.

Nevertheless, the real data exists, and reveals a very different reality from the dominant narrative.

Overall expenditures (including public and private insurers) on mental health in America increased from $32 billion in 1986 to $186 billion in 2014. Even adjusted for inflation, that’s approaching a tripling in funding. (And it doesn’t include substance use treatment beds and services, which also ballooned.) Compared to the 1955 budget of about $364 million for the psychiatric patients in all state and county hospitals, the increases are even more staggering: By 2014, America’s population had doubled while its inflation-adjusted spending on mental health services had gone up 58 times.

In tandem, where comprehensive bed numbers can be found, there are up to two times or more psychiatric beds per person as existed in the 1950s—and most of these beds can or do involve coercive care.

Involuntary psychiatric detentions in inpatient facilities have been rising dramatically. For example, even with only partial data since 2010, UCLA researchers recently found per-capita detention rates across 22 states rising at three times population growth. At 357 per 100,000 people, the U.S. detention rate is double, triple, or many more times the rates documented in the U.K. and comparable Western European countries. Where older data can be found, such as in Florida, these upward trends have been going on for decades.

Meanwhile, many community beds are linked to ACT or similar case-management teams. As I examine in my book, even when they’re not administering forced drugging through court-ordered guardianships or Assisted Outpatient Treatment, surveys show ACT teams still tend to be highly coercive, frequently pressuring clients to take psychotropics and getting them kicked out of housing or involuntarily hospitalized if they don’t comply. In California, for example, a study found that the majority of FSP clients must comply with treatments in exchange for either the supports or housing or both, and a report to the state legislature predicted that California’s new coercive CARE Courts will dramatically increase the number of people in FSPs. Research in other states has also shown that the majority of people accessing voluntary outpatient services have experienced threats to remain treatment-compliant or risk losing their housing, income, or other supports.

It appears that there’s a “shortage” of psychiatric beds and involuntary treatment in the same way some people argue that there’s a shortage of prison beds in the world’s most carceral nation. A more reasonable descriptor might be “overloaded.”

Every year, ever more people are being channeled, voluntarily and involuntarily, towards psychiatric diagnoses and interventions. Today, more than 8 million Americans are receiving some level of care in America’s public mental health systems—colossally dwarfing the numbers getting public mental health care in the 1950s. Yet the numbers keep rising. Evidently, many of these patients are not improving—or may be worsening, as mounting evidence suggests, from long-term iatrogenic impacts from treatments—and are then not able, or not allowed, to leave their psychiatric beds behind.

Arguably, many of these people would be better served with more affordable housing, higher social security payments, social supports without psychiatric coercion, less aggressive nuisance-policing, improved work conditions and opportunities, different approaches in schools and foster care to distress and disruption, helplines that don’t initiate involuntary psychiatric hospitalizations, and equity and justice advances in society. Instead, the mental health system is being pressed to “solve” the impacts of innumerable social problems far outside its capacity—and further increasing the reach, frequency, and intensity of psychiatric cajoling and force appears to be the primary tool in the toolbox.

Ultimately, rather than the common characterization of a desperately underfunded, threadbare mental health care system that’s been shrinking since the 1950s and where practically no one gets subjected to involuntary treatment anymore, in fact, America has massive, expanding systems of psychiatric coercion threading their way throughout its communities and institutions.

Even in the APA publication Psychiatric News, psychiatric historian Jeffrey Geller similarly described so-called “deinstitutionalization” as instead a nearly all-pervading “transinstitutionalization” from asylums to a plethora of smaller psychiatric-carceral facilities and coercive community beds and services. Among people labeled with serious mental illnesses today, wrote Geller, the only ones “who may be totally unfettered” from psychiatric coercion are “the homeless.”

In similar recognition of this reality, after extensive consultations with persons with disabilities, the 2022 “Guidelines on Deinstitutionalization” from the United Nations Convention on the Rights of Persons with Disabilities declared that, “Housing should be neither under the control of the mental health system… nor conditioned on the acceptance of medical treatment or specific support services.”

Moving ahead, then, let us at least accurately describe what has been happening: America’s mental health system has never had more funding, more psychiatric beds, or more coercive treatment—and to the extent the system is failing, THAT is the approach that is failing.

***

MIA Reports are supported, in part, by a grant from The Thomas Jobe Fund.

***

Appendix: Sources for Estimated Numbers of Other Types of Beds

Below are explanations and reference links examining estimates of the numbers of beds in some of the other main types of inpatient and community residential psychiatric beds that were left uncounted in the NASMHPD, SAMHSA, and APA reports.

Child and youth inpatient and residential beds

NASMHPD report author Ted Lutterman’s count included about 42,000 children and youth he found in inpatient psychiatric facilities on a single day in 2018; however, he described and confirmed to me in discussion that many other child and youth psychiatric beds likely exist outside common tracking systems—including in the so-called “troubled teen industry” and out-of-state institutional foster-care settings. Indeed, in an estimate still widely cited today, a 2008 Government Accountability Office (GAO) investigation found 200,000 children and youth in such inpatient-like residential psychiatric institutions receiving federal support, plus an “unknown number” placed “by parents or others” in psychiatric facilities that were often doubling as boarding schools, academies, or boot camps. Because these are residential settings, stays are typically for long periods—so if there were in fact 158,000 or more missed patients in Lutterman’s count, the number of missed beds might be equal to that or more.

Scatter beds

About 80 percent of general hospitals—over 4,000 hospitals—do not have specialized psychiatric wards, but do have “scatter beds” for psychiatric care. These are general beds that sometimes get used for psychiatric care. No one tracks these, but a 2010 study cited by Lutterman estimated a significant 6 percent of all general hospital psychiatric patients stay in these beds—by 2014, that would have been about 121,000 people yearly. Some older studies put the estimate five times higher.

Other nursing home beds

The numbers of nursing home beds holding people diagnosed with depression or anxiety disorders is between 616,000 and 988,000 (there are probably people labeled with both disorders). Lutterman excluded these people because many likely have dementia or serious physical debilitation—yet there’s no data on what percentage are primarily psychiatric patients, though such patients are frequently placed in nursing homes. Given the large numbers, even a small percentage could be a significant number of additional beds.

Emergency room beds

A reported 70-80 percent of hospitals “board” psychiatric patients in their emergency rooms for hours, days, weeks, or even months at a time. The practice has been growing and, in a 2015 study using 2008 data, 1.7 million psychiatric patients annually were boarded in emergency rooms. Many such patients get subjected to forced psychiatric interventions.

Psychiatric beds in prisons and jails

Lutterman noted that many prisons and jails actually have inpatient psychiatric units, but the numbers are currently unknown. In effect, these are the equivalent of high-security state hospital beds that, from the 1950s until today, have been used for criminal/forensic psychiatric patients. On top of, or including those, there are the equivalent of publicly funded residential (albeit carceral) psychiatric beds. A 2016 lawsuit revealed that 32,000 prisoners were getting regular psychiatric treatment in California alone, both voluntarily and involuntarily, and a 2017 lawsuit firmly established a range of psychiatric services for them. Extrapolated nationally, this could be 288,000 beds.