Olivia de Havilland and Betsy Blair as inmates at a mental hospital in The Snake Pit, 1948

20th Century Fox/Everett Collection

Olivia de Havilland and Betsy Blair as inmates at a mental hospital in The Snake Pit, 1948

Although few people are satisfied with the quality of mental health services in the US, it is still startling to find physicians and psychiatrists enthusiastically calling for a return to asylums. One might think that the grim history of confinement would have precluded such advocacy. Whether in popular imagination (think of The Snake Pit or One Flew Over the Cuckoo’s Nest), historical literature (Michel Foucault’s Madness and Civilization), or legislation (exemplified by the Americans with Disabilities Act), the asylum seemed to represent yesterday’s nightmare.

Nevertheless, last winter, The Journal of the American Medical Association ran an editorial titled in part “Bring Back the Asylum,”1 and a few weeks later, a New York Times Op-Ed piece endorsed it.2 In this same spirit, Massachusetts recently invested $300 million in a new mental hospital, the Worcester Recovery Center and Hospital. Are the failures of community services so extreme and beyond repair as to justify recreating asylums?

The bleak history of asylums is worth recalling. In the opening decades of the nineteenth century, widespread and exaggerated optimism about the ability of well-ordered institutions to cure insanity led state after state to build asylums. Initially, they were points of pride, with some claiming 100 percent cures, but within forty years, they deteriorated into custodial facilities. Untrained and unsupervised staff stood guard over patient populations that were predominantly composed of immigrants, many of them elderly. Periodic exposés documented gross overcrowding and mistreatment but did not lead to corrective action. Legislatures, content to rid the streets of troublesome individuals, kept their appropriations meager.3

During the first half of the twentieth century, state hospitals became more and more overcrowded and understaffed. In New York and Massachusetts, for example, they held patients far in excess of capacity. During 1925, there was an excess of 27 percent in New York and 14 percent in Massachusetts. On average, there was one physician for every 250 patients, so the institutions were almost entirely custodial.4 To make matters worse, they became a preferred site for human experimentation. Superintendents approved the research programs and patients had neither the capacity nor the opportunity to say no; to researchers’ delight, there was no effective oversight of the work. In World War II, mental hospitals were the setting for experiments with malaria and flu; patients were purposely infected with the diseases so that the efficacy of experimental drugs could be evaluated. Immediately after the war, the institutions were used for vitamin experiments. Researchers restricted patient diets to induce vitamin deficiencies so that they could investigate the adverse effects. What better way to learn that without vitamin B, patients developed lesions, physical infirmities, and neurological problems?5

Beginning in the 1960s, a process of deinstitutionalization began, the result of an unlikely combination of factors. One was the advent of new pharmaceuticals that controlled psychotic episodes, thereby enabling some of the mentally ill to stabilize their lives outside the asylums. Second, many states saw a financial advantage in reducing the size of the mental hospital population—the primary goal was not to benefit patients but to reduce expenditures. Not only would mental health agency budgets shrink but, if discharged patients were placed in private nursing homes, federal Medicaid dollars would share the costs.

Ronald Reagan was particularly successful in implementing this strategy in California. Indeed, Reagan further benefited because some proprietors of the nursing homes, which were often for-profit institutions, became financial contributors to his political campaigns. States promised to deliver more mental health services to mentally impaired people in these communities; but as we shall see, this promise was not often kept.

Third, and perhaps most essential to the deinstitutionalization movement, was the litigation brought by civil liberties lawyers. Some became involved because of their concerns for patients who wanted to challenge their compulsory confinement in institutions or who wanted to resist the forced consumption of psychoactive drugs or regimens of electroshock therapy. Some lawyers were appalled by the wretched conditions in state mental hospitals and the even worse conditions in facilities for the mentally disabled. In 1967, the New York Civil Liberties Union established a pioneering project on Civil Liberties and Mental Illness. In 1973, the American Civil Liberties Union, the Center on Law and Social Policy (a Washington- based public interest group), and the American Orthopsychiatric Association created the Mental Health Law Project (MHLP), aimed at helping people living in wretched conditions, many in state institutions.

One lawyer, the late Bruce Ennis, became a central figure in establishing the legal standards both for involuntary confinement and for conditions of confinement. Among his most noteworthy cases was O’Connor v. Donaldson, which he argued before the Supreme Court in 1975. The plaintiff, Kenneth Donaldson, a carpenter, had been committed to a Florida mental hospital after his elderly father petitioned a state court to hold a sanity hearing for his son. The finding was that Donaldson suffered from paranoid schizophrenia; after a very brief hearing, the judge remanded him to the Florida State Hospital, an institution with some five thousand inmates. Donaldson rarely saw a physician there during fifteen years of confinement. When he did, he claimed, the encounters were brief. Typically, the physician asked three questions: “What ward are you on?” “Are you taking medication?” “Are you working any place?” “That will be all.”


Wanting his freedom, Donaldson sent many petitions to courts; one of them succeeded in his being given a hearing, and the case wound its way to the Supreme Court. The institution defended itself by arguing that Donaldson had received “milieu therapy,” that is, he had benefited from confinement in the mental hospital. Ennis argued for him, and the Supreme Court decided unanimously that “a State cannot constitutionally confine, without [treatment], a nondangerous individual who is capable of surviving safely in freedom by himself or with the help of willing and responsible family members or friends.” Thus, state authority to decide who could be confined to a mental hospital began to be circumscribed.

Another case that Ennis argued, Wyatt v. Stickney (1971), before federal judge Frank Johnson in Alabama, helped define the responsibilities for treatment in state mental hospitals. Patients could not simply be kept under lock and key. They had a legal “right to treatment,” and courts were duty-bound to enforce the standard. Were it not met, the courts were to consider the ways that patients could be released. Ennis also helped bring judicial standards to the state facilities for the mentally disabled. His suit against the Willowbrook State School, on New York’s Staten Island, culminated in a consent decree under which the state agreed to resettle the residents in group homes, with the entire process under court supervision.6

The advocacy of organizations like the MHLP, renamed the Bazelon Center (in honor of David Bazelon, a federal judge who vigorously promoted the rights of the mentally ill), also contributed to the adoption of major federal legislation, most notably the Americans with Disabilities Act of 1990 (ADA), which articulates the values and practices aimed at expanding the life chances of the physically and mentally handicapped. The bill expresses a strong preference that services and accommodations be provided for the disabled in the community, with minimal restrictions and regular contact with people who are not disabled. As the act notes:

Historically, society has tended to isolate and segregate individuals with disabilities, and, despite some improvements, such forms of discrimination against individuals with disabilities continue to be a serious and pervasive social problem.

The bill also instructs the US attorney general to issue regulations that provide for services in “the most integrated setting appropriate to the needs of qualified individuals with disabilities.” The regulations carrying out the act define such a “setting” as one “that enables individuals with disabilities to interact with non-disabled persons to the fullest extent possible.”

The leading Supreme Court decision interpreting the ADA requirements is Olmstead v. L.C. (1999). The case involved two women in Georgia, L.C. and E.W., both identified as mentally retarded, or in current usage, “developmentally disabled.” L.C. had also been diagnosed with schizophrenia and E.W. with a personality disorder. Both women had been confined at the Georgia Regional Hospital in Atlanta. The psychiatrists treating them had determined that placement in a community-based program would be appropriate for them, but their confinement continued. They then filed suit and asked to be moved to a community care residential program, with the ultimate goal of integration into mainstream society.

Writing for the court majority, Justice Ruth Bader Ginsburg agreed, asserting that under the ADA,

States are required to provide community-based treatment for persons with mental disabilities when the State’s treatment professionals determine that such placement is appropriate, the affected persons do not oppose such treatment, and the placement can be reasonably accommodated, taking into account the resources available to the State and the needs of others with mental disabilities.

As is apparent, this was hardly a radical holding. It did not require that people be released from institutions over the objections of treatment professionals or patients; also, such release was made contingent on the resources available to the state. Justice Ginsburg’s opinion pointed out that a lower court had found that community care in local living facilities is less costly than institutional confinement; but the decision also recognized that, in some cases, a state might face increased costs by funding placements in community facilities before being able to reap the savings involved in closing an institution.

There is no denying that despite the best efforts of lawyers and advocates, dollars and services have not generally followed patients released from institutions into community facilities. The mentally disabled have fared better than the mentally ill, first because parents have been relentless supporters of the rights and needs of their children. They discover the handicap of their child soon after the birth and become ardent advocates for programs and services. Second, Medicaid funding for the mentally disabled—but not for the mentally ill—is increasingly available to encourage release from institutions and the phasing out of the institutions themselves. States can draw on Medicaid matching funds (through the so-called ICFMR regulations) to support community-based facilities that provide residents with “active treatment.”


The outcome for patients released from mental hospitals—or, for that matter, community residents in need of assistance—has not been nearly as favorable. Before the Affordable Care Act (ACA), federal dollars were not available for adults below the age of sixty-five to obtain community mental health services. Moreover, there is no lobby for the mentally ill comparable to the lobbies of parents of the mentally disabled. Mental illness, which often tends to manifest itself in late adolescence or early adulthood, may lead to an estrangement from family. Whatever the reasons, state commissions and investigative reporters provide painful evidence of maltreatment or no treatment at all for the mentally ill.

New York, for example, placed (at a cost of $40,000 a year) discharged mental patients into so-called adult homes, typically run by for-profit companies; the homes were supposed to provide a safe environment and access to treatment, but they did neither. As Clifford Levy, a journalist for The New York Times, reported in a series of Pulitzer Prize–winning articles in 2002, the facilities were typically run-down former hotels or office buildings, three to five floors, with anywhere from fifty to two hundred rooms that residents usually shared.7 They included a communal kitchen, dining room, and smoking room, described by Levy as a “cluster of worn benches, bare walls,…and a floor littered with cigarette butts, spilled coffee, ashes and discarded food.” Staff were poorly paid and untrained, incompetent or unwilling to arrange for the psychotherapy that residents wanted and needed.

‘Dance in a Madhouse’; lithograph by George Bellows, 1917

Metropolitan Museum of Art

‘Dance in a Madhouse’; lithograph by George Bellows, 1917

This persistent failure to deliver decent and effective community care also reflects the shortcomings of state agencies. Again, the New York experience is instructive. It took some twelve years and extensive litigation to get New York to agree to assess the ability and willingness of four thousand residents of adult homes to move into private apartments (so-called supported housing units, which are intended to allow residents to obtain appropriate treatment). But the actual pace of change has been sluggish. The federal judge who demanded and approved the settlement, Nicholas Garaufis, has complained about how slowly necessary changes are being made: in his opinion, delivered in 2014, he wrote, “Relief for these residents is long overdue, and the court considers this delay to be unconscionable.”8

But even he and the court monitor he appointed, Clarence Sundram, an experienced advocate on behalf of the rights of the mentally disabled and the mentally ill, for all their oversight, have been unable to expedite the process. The reasons for the state’s delays are telling. The agreement that 4,197 residents of adult homes would be able to move into private apartments went into effect on July 23, 2013. Between then and March 13, 2015, 1,256 of them were interviewed and 758 wanted to move into private apartments. But the number who actually moved was only forty.

Why so few? Some residents changed their minds about moving, because self-interested staff at adult homes persuaded them or their families that they should stay put. Other residents wanted to move to neighborhoods where housing costs exceeded state budgetary allotments. There were lags in the required assessments of individual cases, which included an updated psychiatric evaluation. There were also difficulties in locating patient records, pervasive staff shortages, and untrained personnel. It was hard to schedule assessments because the people employed by the adult homes were uncooperative, reluctant to complete evaluations that would enable residents to move out. In effect, the state could not get its act together even to fulfill a court order it had accepted.9

Finally, all these failures are compounded by the fact that a significant number of former mental patients end up in local jails, often for petty offenses such as disorderly conduct. There they join others who were never committed but also suffer from mental illness. A recent report estimates that nearly 40 percent of the inmates of New York’s Rikers Island jail, where the costs of confining them substantially exceeds those of adult homes, have some form of mental illness.10 Cook County Sheriff Thomas Dart recently commented:

The vast majority of mentally ill people are here for nonviolent crimes, like stealing food to survive or breaking into places, usually looking for somewhere to sleep, or getting caught with drugs because they are self-medicating. How is it different than us locking up diabetics? Jails were never meant to be mental health hospitals.11

Still, returning to the days of the asylum would be an egregious mistake. As daunting as it is to assure quality of care in community settings, it is far more difficult to assure that people will be treated decently in closed institutions. To assert, as the authors of the JAMA commentary did, that the asylum was “a protected place where safety, sanctuary, and long-term care” was provided and that “it is time to build them—again” is to turn a blind eye to the established facts of asylum life. To give asylums priority over community settings would send the unmistakable message that the disabled are first and foremost a public hazard, best managed by incarceration. As stingy as legislators may be in providing for facilities in communities—whether in families, adult homes, or private apartments—they are still more stingy in appropriations for sealed-off institutions. Out of sight is out of mind. The likelihood that asylums will be adequately funded and monitored is low, and none of their proponents has suggested policies that might produce a different outcome.

Even more troubling, the advocates of asylums completely fail to take into account the many provisions of the Affordable Care Act that have the potential to transform the care of the mentally ill—both those who have been institutionalized and those who have not.12 The act defines ten “Essential Health Benefits” that must be part of insurance coverage plans, and they include not only ambulatory care, emergency care, and hospitalization but also “mental health and substance use disorder services, including behavioral health treatment,” that is, community care. In addition, the ACA health insurance exchange provisions give subsidies to low-income people, which will include many of those in need of mental health services. (Some 3.7 million heretofore-uninsured people are estimated to have severe mental illness.) The ACA also requires parity in insurance coverage between physical and mental health services, prohibits discrimination for preexisting medical conditions, and extends coverage for children under parental policies to age twenty-six. All these provisions will expand the insurance available to people in need of mental health services.

Beyond all this, the ACA expands the scope of Medicaid benefits for mental health, providing incentives for health care organizations to combine their services. With substantial federal funding, at least at the outset, providers can establish “medical homes” and “accountable care organizations” to coordinate diagnoses and treatment for both physical and mental health conditions. They would be in a position not only to treat patients but also to assist them in obtaining legal advice, housing, and employment—one-stop shopping for medical and social services. The so-called Medicaid 1915(i) state waiver option expands states’ abilities to provide benefits at home and in community clinics to specific groups of people, including the severely mentally ill. The ACA even has funds for a “Community First Choice” program, giving states the wherewithal to provide low-income people with local services expressly to prevent confinement in institutions.

There is no certainty that states will take advantage of these new incentives and opportunities. Few states have chosen to participate in Community First Choice programs and twenty-two states have refused to expand eligibility for Medicaid programs. So too, insurance companies may find ways to avoid paying for mental health treatment—a charge already being made. Medical homes and accountable care organizations, provided for by the ACA, are just getting underway and those in charge will have much to learn if they are to be effective. But how much better to have new prospects for keeping people at home and in their communities than to issue tired calls for more asylums.

Finally, a commitment to realize the principles of the ADA may be reinforced by current developments in criminal justice.13 Both liberals and conservatives are recognizing that the US is incarcerating far too many people for too long at enormous financial cost. The US prison population has increased fivefold since the 1970s. This is partly the result of over-incarceration of nonviolent and low-level drug offenders, and of sentences that are many times longer than they are in European countries. It is also the result of three-strike laws, and the virtual elimination of parole.14

President Obama has recently made reducing prison populations a political priority, and even the Koch brothers have joined with such philanthropists as George Soros in support of such efforts. Congressional approval might be forthcoming. States that have declined to take federal funds for Medicaid expansion may recognize that their own expenditures on correctional institutions will remain high unless they find ways to reduce the number of mentally ill in jails and prisons. The community services that Medicaid helps to support would go a long way to accomplishing this goal. The policies that would emerge—such as expanded programs to integrate offenders back into the community—could also prove highly beneficial for the disabled. In the end, it is in everyone’s interest—offenders, disabled, and the general public—to have fewer people under lock and key.