My brother Robert, for whom I was advocate and caregiver, spent more than fifty years in and out of facilities where he was overmedicated, beaten up by aides, and frequently put in isolation—which is to say left, sometimes for several days, in a small room with only a sheetless iron bed, a treatment the staff called “reduced stimulation.” Such “treatments,” in what are now called psychiatric centers but were previously called lunatic asylums, left him with lifelong side effects, including drug-induced parkinsonism, diabetes, congestive heart failure, and the loss of all his teeth.
In the 1990s, when Robert was in his fifties, he lived for two years at the Bronx Psychiatric Center in New York City, where Oliver Sacks had earlier worked for more than two decades. “As at all such hospitals,” Sacks wrote in a 2009 essay in these pages, the facility had “great variations in the quality of patient care: there were good, sometimes exemplary wards, with decent, thoughtful physicians and attendants, and bad, even hideous ones, marked by negligence and cruelty.” Robert’s was one of the exemplary wards, the only training ward at the hospital, with a staff that included psychiatric residents and psychology interns.
When Robert arrived there, as I noted several years ago in an article for the Wall Street Journal, the staff was convinced that he would never be able to live outside an asylum. But the director of psychology, Alvin Pam, disagreed. Under Doctor Pam’s supervision, Robert’s treatment included an hour a week of one-on-one psychotherapy. Within two years he recovered well enough to be discharged to an excellent halfway house in Hell’s Kitchen, and for the next dozen years he enjoyed a new measure of happiness, stability, and independence.
Today many of the institutions through which Robert passed are closed, or greatly diminished. In 1955 the United States had 352 state psychiatric institutions housing 558,992 patients. It now has 182, which house fewer than 40,000. This dramatic change had its origins in the movement for deinstitutionalization in the 1950s and 1960s, when critics of asylums proposed replacing them with community mental health centers that, they believed, would enable people with mental illness to recover without being displaced from their neighborhoods, families, and familiar support networks. In 1963, with President Kennedy’s endorsement, Congress passed the Community Mental Health Act, which called for vastly reducing the use of state hospitals.
It said nothing, however, about where the half-million discharged people would live. Little of the money saved by closing hospitals followed the patients. The new community centers provided neither housing nor rehabilitation nor medical care; in practice, according to Michael B. Friedman, a former deputy commissioner of mental health for New York State, they treated mostly “people with less severe mental disorders” rather than those who had been discharged from asylums. By the mid-1980s, meanwhile, low-income housing was becoming increasingly scarce, and fewer vulnerable people could afford what options the market offered; thousands, for instance, lost their Social Security Disability Insurance benefits in the first half of the decade when the Reagan Administration instituted stricter eligibility criteria.
Once the process of deinstitutionalization got underway, public hospitals started allocating fewer beds to patients struggling with mental illness. In 1955 half of all public hospital beds in the US were psychiatric; today we have eleven psychiatric beds for every 100,000 people. (Countries such as Belgium, Germany, Norway, and Latvia have more than a hundred.) In New York City the number of inpatient psychiatric beds has fallen by 20 percent since 2014, mostly because, according to unnamed state officials quoted in The New York Times, it became “more profitable for hospitals to redirect resources elsewhere.”
The number of homeless people declined slightly in the years before the Covid-19 pandemic, but in 2022–2023 the national homeless population rose by a rate of 12 percent, even as the number of available psychiatric beds was shrinking. The following year it rose by 18 percent. On the coasts that rate is much higher: between 2014 and 2020 California’s homeless population increased by 42 percent, and between 2022 and 2024 New York’s rose by 53 percent. The Department of Housing and Urban Development estimates that more than 770,000 people in the country are homeless, more than half of them on the coasts.
According to Deborah K. Padgett, a professor of social work at NYU and former president of the Society for Social Work, “epidemiological studies have consistently found” that 25 to 30 percent of that population struggles with serious mental illness. The longer they remain homeless, the more severe and intractable their conditions become and the likelier they are to get caught in a cycle of “transinstitutionalization,” moving from crisis hospitalization to homelessness to shelters to incarceration and back again. In cities such as Los Angeles, Chicago, and New York, they often spend an inordinate portion of their lives behind bars. The New York State Bar Association notes that “people with mental illness in the US are ten times more likely to be incarcerated than they are to be hospitalized.” As many as “a third of incarcerated people,” it reports, “have a serious mental illness.” The nation’s two largest mental hospitals are, effectively, located within prison facilities: Rikers Island in New York City and Twin Towers in Los Angeles.
*
In recent years a range of politicians and other public figures—some spurred by tabloid-like reports of disruptive and violent behavior by individual homeless people, others by a desire to help such people recover and lead viable lives—have responded to this trend by calling not for revitalizing community mental health centers but for reestablishing asylum-like facilities. In 2015, for instance, the bioethicists Dominic Sisti, Andrea Segal, and Ezekiel Emanuel coauthored a paper in the Journal of the American Medical Association called “Improving Long-Term Psychiatric Care: Bring Back the Asylum.”
The “original meaning of psychiatric ‘asylum,’” they wrote, was “a protected place where safety, sanctuary, and long-term care for the mentally ill would be provided. It is time to build them—again.” The alternative was bleaker, they argued: “at the moment prisons appear to be the default option.” Versions of that argument have been taken up more recently by other prominent figures in the field, including the NYU bioethicist Dr. Arthur Caplan and the psychiatrist Dr. E. Fuller Torrey, who founded a nonprofit called the Treatment Advocacy Center in 1998 and has long criticized the movement for deinstitionalization.
Writing in 2023 in The Wall Street Journal, the Pulitzer Prize–winning historian David Oshinsky spelled out the case at some length. “For the most vulnerable among us,” he suggested, reviving state-run mental institutions would be “a far better option than the alternatives of homelessness and incarceration.” Oshinsky acknowledged that asylums have a far from reassuring track record. He had in mind, he stressed, something other than “torture chambers where icepick lobotomies, electric shock, sterilization and solitary confinement turned humans into zombies.” As a model for ending “the revolving door of homelessness-hospitalization-prison that passes for policy today,” he offered the same facility that Emanuel and his coauthors had: the Worcester Recovery Center in Massachusetts, “a facility for 320 long-term patients, with private rooms.” Oshinsky’s essay ran under the headline “It’s Time to Bring Back Asylums.”
Online, a number of residents do indeed praise the state-run Worcester center—opened in 2012 on the grounds of an abandoned asylum—for the sense of safety and security it provides, for the wealth of activities it offers, for its devoted therapists and multiple therapeutic options, and for the fact that each patient has their own room and shower. These services did not come cheap: in a response to Oshinsky, Steven Mintz notes that the center cost $300 million to complete and spends $200,000 annually per patient.
Could such a facility be replicated in the US at the necessary scale? In recent years several states, including Texas, Indiana, North Carolina, Michigan and Missouri, have indeed invested substantial sums in the construction and renovation of mental health facilities, most of them on the sites of old state asylums. Texas, for example, has appropriated more than $2.5 billion “to replace, renovate and expand state hospitals”—a welcome move in a state that in 2022 ranked last in the country in access to mental health care. According to the state’s executive commissioner of Health and Human Services, the first completed new facility, the Dunn Behavioral Sciences Center in Houston, “accommodates multiple patient populations and levels of treatment with access to medication management, group and individual therapy, and educational and life skills training.”
These new and renovated facilities hope to provide much-needed and improved mental health services to previously underserved populations. None of them, however, offers enough services to address the needs of the large and growing number of homeless people with serious mental illness within their geographic purview. The Dunn center has 264 beds in a metropolitan area where, as of last year, more than 1,100 homeless people suffer from serious mental illness. In Missouri more than a thousand homeless people meet the same criterion; the state’s flagship new mental hospital, which cost $211 million, has beds for three hundred of them.
According to HUD, the homeless populations of California and New York add up to more than 345,000 people. If only 25 percent of this population suffers from serious mental illness, then those two states alone would require psychiatric centers capable of offering long-term housing and treatment to more than 86,000 patients. The entire country, by the same rough calculation, would need centers capable of housing and treating roughly 192,500 people. In a political climate that favors severe governmental cost-cutting, it is hard to imagine states having the resources—or political will—to build anywhere near enough high-quality new psychiatric centers to meet that demand.
Meanwhile the call to bring back the asylums has also been taken up by public figures who have a very different kind of institution in mind—politicians who believe that most homeless people do not deserve the housing, care, and treatment that writers like Oshinsky and Emanuel propose. Donald Trump has long advocated for incarcerating homeless people in asylum-like facilities, sometimes with surprisingly bipartisan support. (“Trump Wants More Asylums,” read one Times headline in 2018, “and Some Psychiatrists Agree.”) He reiterated that view in a 2023 campaign video, promising upon reelection to bring “those who are severely mentally ill and deeply disturbed…back to mental institutions, where they belong.”
Two former Trump administration members have expressed support for bringing back asylums: Robert Marbut, executive director of the Interagency Council on Homelessness, and Ben Carson, secretary of HUD in Trump’s first administration. When he was a lawmaker in the Texas state legislature, Scott Turner, Trump’s current HUD secretary, more generally showed “a deep-seated skepticism about the value of government efforts” to combat poverty and homelessness, according to ProPublica:
Turner supported a bill ensuring that landlords could refuse apartments to applicants who received federal housing assistance. He opposed a bill to expand affordable rental housing. He voted against public-private partnerships to support the homeless and against two bills that called merely to study homelessness among young people and veterans…. He has called welfare “dangerous, harmful” and “one of the most destructive things for the family.”
*
Writers like Oshinsky and Emanuel acknowledge that, for the most part, the old asylums sucked the lives and souls from tens of thousands of sick, vulnerable human beings. Their call for “bringing back the asylum” differs dramatically from Donald Trump’s, but the problems posed by the asylum model are not easily avoided. In the absence of any concrete plan—or even the prospect of a plan—for creating, funding, and maintaining what Sisti calls “high quality, ethically administered psychiatric asylums” at scale, any program that legitimizes removing homeless people with serious mental illness from our communities and putting them somewhere else—whether a hospital, a prison, a jail, or an as-yet-unbuilt asylum—carries large risks. Where and how would one find places sufficient to house nearly 200,000 sick, vulnerable people, much less provide treatment that would give them a chance to recover and return to society as our friends, neighbors, and coworkers? How would one prevent new state mental health facilities from devolving into the very sorts of hellholes these writers oppose—once-humane institutions that devolved into notorious “snake pits” and often tragically blunted their residents’ lives?
Even were it possible to build enough new facilities at scale and keep them from devolving, there is also a persuasive case to be made that the very form of the asylum inhibits patients from achieving genuine recovery. In 2015 Lloyd I. Sederer, former chief medical officer of the New York State Office of Mental Health, offered a salient point. “The competencies we all seek,” he wrote,
cannot be achieved in a long-term institution. Institutions, by their nature, foster helplessness and dependency. We should not regard anyone with an illness, including a mental illness, as hopeless—to be cloistered away until death takes them away. No one should be expelled from home and community for a lifetime of institutional care, which is what a policy of asylum care would mean.
Sederer makes a troubling implication: that the net effect of all these calls to bring back the asylum—even with the best intentions—is to stigmatize people who struggle with mental illness and reinforce the idea that their presence in our communities is disturbing. Even as well-informed and well-intentioned an essay as Oshinsky’s begins with a description of three murders committed by “unsheltered homeless men with violent pasts and long histories of mental illness.”1 Meanwhile, Sederer argues, “romanticizing the asylum…risks diverting attention and vital resources (i.e., money) from the current, mounting efforts to build an adequate mental health system.”
*
What might such a system look like? As Friedman has argued, giving people with serious mental illness a chance to recover from their long-term conditions means giving them access to medical treatment, which in turn, “depends first and foremost on a decent place to live.” For individuals like my brother, who already had a fragile purchase on the world, transient housing and intermittent medical treatment simply do not do the job. Mitchell Katz, president and CEO of New York City’s public hospitals, put it bluntly in a 2023 comment to the Times: “Nothing about a seven-to-fourteen day hospitalization is going to change the arc of the life of somebody who’s homeless and has schizophrenia.”
People with severe mental illness, however, will struggle to find decent, affordable housing as the market shrinks. “New York City’s housing shortage is at its worst point in half a century,” the Times reported last year. “The number of cheap apartments available to rent…is basically zero, according to the most recent city survey.” When rents rise, individuals with histories of serious mental illness are the least able to afford them, the least able to navigate relocation and, invariably, the least desirable tenants. When someone has been homeless or institutionalized for a significant period of time, for that matter, other problems arise: What sorts of care and treatment will they need as they transition into a new life? What about the years they lost? How does one account for those years to oneself and to others? What sorts of ongoing support does one need from competent, caring professionals to deal with the trauma and shame of having spent long periods of time institutionalized, incarcerated, or homeless?
Since the late 1990s organizations that advocate for people who suffer from homelessness and serious mental health issues have piloted a number of programs that attempt to address these challenges outside of the market on the one hand and the asylum model on the other. After Robert was discharged from the Bronx Psychiatric Center, he lived in a Hell’s Kitchen facility run by a nonprofit called Project Renewal that, in addition to housing fifty-seven men and women, provided on-site social workers, a part-time psychiatrist, a nurse, and vocational and employment advisers. After six years there he moved into a nearby brownstone (built in 1887), where he had his own room alongside seventeen other people who had previously lived in supported housing facilities. It was operated by a national organization called Fountain House, which continues to run that residence along with a large nearby “clubhouse” in which people in various stages of recovery from serious mental illness work side by side with staff to run daily operations, receive medical and mental health services, and get support for employment, housing, and education.
Both residences have strong track records at helping homeless people with serious mental illnesses recover into the blessedly imperfect lives most of us lead. Generally termed “supported housing” (or, in New York state, “supportive housing”), they start by simply providing a place to live. Most such housing is operated by nonprofit agencies and has staff trained in social work and psychiatric rehabilitation. They operate on the assumption that as people with serious mental illness get better they will need less intensive levels of care—that they will move along a continuum from homelessness and hospital stays to halfway houses to supported apartments and ultimately to independence and reintegration in the community.
Some of the programs offer permanent housing, some offer transitional housing, and some offer “scattered-site” apartments where residents live in buildings that may have only a few people with histories of being recently homeless. Many programs, like Project Renewal, also offer congregate housing, or group homes, in which all the residents are there because they have serious mental illness and have been homeless in the past. But congregate housing, Friedman has noted, “is not a great model for people with frequent recurrent illnesses,” nor “for grown-ups who don’t want to live like college kids in shared rooms with community meetings.”
Then there is a model known as “housing first.” Its approach is simple: offer homeless people apartments without preconditions. Housing first programs often take people whom municipalities and supported housing programs refuse (because they are not off drugs, for instance, or not taking their meds), gives them scattered-site apartments, and then bring essential services (psychiatric, social, educational, vocational, medical, legal) to them where they live.
I’ve been working since 1999 with Pathways to Housing in New York City, where the “housing first” model originated, and since 2014 with the Housing First Institute, which operates in the US and countries like Finland, Ireland, New Zealand, and the Netherlands to initiate housing first programs and train their staff. These programs provide members with long-term apartments without requiring that they pass sobriety tests, or show up for appointments at clinics or other offices. By putting housing before treatment—without requiring individuals to “earn” the right to an apartment by moving along a continuum of increasingly less restrictive placements—they enable formerly homeless people to think beyond survival: to think instead about the next hour, or day, or month in their lives.
These programs cost significantly less than other alternatives. A 2021 report by the New York City comptroller Scott Stringer found that the city spends $556,539 on every incarcerated person per year—a figure that has quadrupled since 2011. The average cost of institutionalizing a person in a state mental hospital varies widely across the country, but Sam Tsemberis, the founder of Pathways to Housing and the Housing First Institute, estimates that it runs between $150,000 and $300,000. (My independent research arrives at a similar approximate range.) When we talked earlier this year, Tsemberis told me that well-staffed “housing first” programs cost, depending on location, and including mental health services, somewhere between $20,000 and $40,000 per person per year.
They boast impressive results. In Milwaukee, between 2016 and 2022, such programs reduced street homelessness by 92 percent; in Houston, from 2012 to 2023, they did so by 64 percent. A national Veterans Affairs-HUD program that used the model reduced a homeless population of at least 65,000 by more than half. During my time at Pathways to Housing and Housing First, more than 80 percent of the thousands of chronically homeless men and women we worked with transitioned to stable housing. I have seen mothers reunite with children the legal system had taken from them; people get off drugs and alcohol with the help of harm-reduction programs; individuals who had been living in burned-out buildings, subways tunnels, and abandoned cars get jobs, go back to high school or college, and graduate.
Working with those families, I have often thought back to Robert’s first hospitalization in an asylum—Creedmoor Psychiatric Center—when he was twenty-three years old. Creedmoor is a New York state facility in Queens that had more than seven thousand residents when he lived there. In my conversations with those residents and their families during the six to seven years Robert spent at the facility in the late 1960s and early 1970s, many told me they had gone long stretches—in one case nearly two decades—without ever seeing a psychiatrist, psychologist, or social worker.
In the next three decades—until the dozen years that followed his stay at Bronx State—Robert was hospitalized for various periods of time in more than a half-dozen other asylums. Whenever we talked about these places, he’d always say the same thing: that his greatest fear was not having another breakdown—his psychotic episodes were usually floridly manic though not, he’d say, without their pleasures—but being locked up again at Creedmoor. Often the mere mention of the name would bring on tremors. “You don’t know, Jay,” he’d say. “You don’t know…you don’t want to know….” Now that we have proven ways to house and care for people with serious mental illness, why ever risk bringing places like Creedmoor back?
English (US) · 