The emptying of our public psychiatric hospitals has been the second-largest social experiment in twentieth-century America, exceeded only by the New Deal. The experiment, undertaken upon remarkably little data and a multitude of flawed assumptions, has received virtually no formal evaluation or assessment to ascertain whether it has worked. Once the spring of deinstitutionalization was wound, it just kept going and going and going. And it continues today—disastrously.

It is important to realize the magnitude of this experiment. In 1955 state psychiatric hospitals housed 558,239 seriously mentally ill persons. If the same proportion of Americans were hospitalized today, when the U.S. population is much larger, these hospitals would contain some 900,000 seriously mentally ill individuals. In fact the actual number is less than 70,000, meaning that the net deinstitutionalization amounts to some 830,000 people—more than the population of Boston, Baltimore, or San Francisco.

Some of those who were released fared better than in the state hospitals of the 1940s and early 1950s: most of these institutions were little more than human warehouses. Researchers had yet to develop antipsychotic and antidepressant drugs, and the hospitals had virtually nothing to offer patients except custodial care. That has changed dramatically in the intervening four decades. We now have an armamentarium of effective psychiatric drugs, and many state psychiatric hospitals have improved markedly. A few, such as the exemplary New Hampshire State Hospital in Concord, provide better care than most private psychiatric hospitals.

Many who were deinstitutionalized, however, are worse off than if they had remained in the hospital. They can be found talking to themselves in public streets and parks, living in cardboard boxes or subway tunnels beneath the city in the middle of winter, or escaping the cold in public libraries. They often end up victimized or in jail, charged with misdemeanors. Hundreds of thousands of the deinstitutionalized mentally ill have died prematurely from accidents, suicide, or untreated illnesses. All too frequently, the consequences of this failed social experiment have been tragic and fatal.

New York State is a typical example of deinstitutionalization, with a net effective deinstitutionalization rate of some 90 percent of all patients. Twenty-nine states have deinstitutionalized a higher percentage, led by Rhode Island with a rate of almost 99 percent. Dr. Seymour Kaplan, one of the pioneers of deinstitutionalization in New York State, later considered it the gravest error he had ever made.

From the earliest days of deinstitutionalization it was clear that many persons discharged from New York's psychiatric hospitals were living marginal lives at best. A 1974 study found that one-quarter of the those living in New York City's welfare hotels had once resided in psychiatric hospitals. By 1979 shocking newspaper articles had grown common. One typical story, headlined "21 Ex-Mental Patients Taken from 4 Private Homes," described the former patients as living in Queens "amid broken plumbing, rotting food, and roaches." The article further reported: "Last May the police found the decaying corpse of a former patient lying undisturbed in one home inhabited by six other residents."

New York State also provides abundant examples of the failures of deinstitutionalization, as measured by the number of mentally ill individuals who are homeless or in jail, the number of violent acts by mentally ill persons who are not receiving treatment, and the overall quality of life in the community.

A 1985 front-page story on the homeless in the New York Times featured a photo of a man named Vito, "who wears a suit and lives in a coffinlike cardboard box." It also described Antoine, who lived beneath FDR Drive, where he was training "for the rigors of a mission to outer space," and a homeless man who lived near the World Trade Center "because he believed it would be a good place to take off for outer space." As recently as 1995 a New York study reported that 38 percent of discharged psychiatric patients "have no known address within six months of their release."

Perhaps the ultimate symbol of the relationship between psychiatric hospitalization and homelessness in New York is the Keener Men's Shelter on Ward's Island. For three-quarters of a century the building was part of Manhattan State Hospital, serving individuals with severe psychiatric disorders. In 1981, as the state emptied the hospital under the banner of deinstitutionalization, the building became a men's public shelter. When I last visited the shelter in 1990, it housed 800 homeless men. Approximately 40 percent of them were severely mentally ill; several had once been hospital patients in the same building. Despite humane management by the Volunteers of America, the Keener shelter is merely a shelter and is neither staffed nor funded to provide the intensive psychiatric care that many of its residents need, including medication maintenance, vocational rehabilitation, social reeducation, and the transition to supported housing of their own or supervised group homes.

Another disturbing legacy of deinstitutionalization is the large number of the mentally ill in New York's jails and prisons. In 1987, Dr. Henry Steadman and his colleagues published the results of interviews with 3,332 state prison inmates and reported that 8 percent of them had "very substantial psychiatric and functional disabilities that clearly would warrant some type of mental health service." The consequences of jailing such people are often fatal. A study of inmates who committed suicide in New York State jails between 1977 and 1982 found that half of them had been previously hospitalized for a mental disorder.

Most of the mentally ill who end up in jail have committed misdemeanors like disorderly conduct or trespassing. For many of those charged with more serious crimes, there is a direct, and usually obvious, connection between their condition and their crime. For example, in Rochester a mentally ill man robbed a bank using his pointed finger in his pocket as a "gun," then took the money to the zoo and threw it into the enclosure for seals, "urging the animals to return home," according to a New York Correctional Association report. The report concluded that such acts were "more [the] pathetic expressions of mental illness than deliberate, premeditated crimes."

A small number of untreated mentally ill individuals, however, do commit serious crimes, including murder. Almost invariably, they are not receiving medication for their illness at the time of the crime. The incidence of such violent acts has grown in tandem with the number of untreated mentally ill individuals.

In New York such crimes have become a continuing commentary on the perversity of deinstitutionalization. Mary Ventura, three weeks after being discharged from a psychiatric hospital, pushed a woman she did not know into the path of a subway train Reuben Harris, with 12 previous psychiatric hospitalizations, pushed another woman to her death in the same manner. Juan Gonzalez, who had been psychiatrically evaluated four days earlier, killed two and injured nine others on the Staten Island Ferry. Kevin McKiever, well known to the city's psychiatric services, stabbed to death a woman who was walking her dogs in Central Park. Christopher Battiste, psychiatrically treated at city hospitals twice in the previous two months, bludgeoned to death an 80-year-old woman on the steps of a church. Jorge Delgado, previously hospitalized seven times for his paranoid schizophrenia, ran naked into St. Patrick's Cathedral and killed an elderly usher. Dennis Sweeney, Lois Lang, Van Hull, Michael Vernon, Steve Smith, Da Pei Wu, Tatiana Belopolsky . . . the tragic litany continues year after year. There are an estimated 1,000 such homicides in the United States per year—4 to 5 percent of all homicides. The common denominator is a severely delusional mentally ill person not receiving treatment.

Finally, of course, deinstitutionalization has caused the quality of life for everyone in the community to deteriorate. Violent acts like those I've mentioned contribute to this deterioration, as noted by a New York Times columnist in 1995 in response to the Reuben Harris case: "If you are a New Yorker the fear is there somewhere, maybe buried deep beneath the surface of consciousness, or maybe right out there in the open. . . . The fear is that from out of the chaos some maniac will emerge to pointlessly, stupidly, inexplicably hurl you, blast you, cast you into oblivion."

But it is not only episodes of violence that sap community life under deinstitutionalization. It is Joyce Brown, mentally ill and addicted to drugs, living on a steam grate at the corner of East 65th Street and Second Avenue, urinating on the sidewalk and defecating in the gutter as she hurls obscenities at passersby. It is Larry Hogue, the "wild man of West 96th Street," with manic-depressive illness and cocaine dependence, masturbating in front of children and threatening to eat the dog of one local resident. As a woman writing a letter in the Wall Street Journal in 1994 summarized it: "A simple visit to the local elementary school, post office or grocery store . . . can be a Dantean journey through the dark underside of our society."

It is important to realize that the original underpinning for deinstitutionalization was political correctness, not scientific knowledge. When deinstitutionalization shifted into high gear in the early 1960s, only one study had been done on the effects of moving severely mentally ill individuals from public psychiatric hospitals to community living. Published in England in 1960 by Dr. John Wing, the study found that 20 schizophrenic individuals, selected because they were functioning at a high level and were able to work, did relatively well when moved from a psychiatric hospital to supervised community living facilities. Virtually every American advocate for deinstitutionalization in the 1960s and 1970s cited this lone paper, despite the fact that the 20 patients studied were far from representative of the vast majority of patients who were sent packing.

The policy of deinstitutionalization was in fact based on ideology. In 1961 psychiatrist Thomas Szasz published The Myth of Mental Illness, alleging that there really was no such thing as schizophrenia or manic-depressive disorder. That same year sociologist Erving Goffman published Asylums, in which he asserted that most of the behavior psychiatric patients exhibited was a consequence of their being in the hospital, not of their illnesses. The following year Ken Kesey published One Flew over the Cuckoo's Nest, which became an underground best-seller and further promoted the idea that most patients in state psychiatric hospitals were characters like Randle McMurphy, who had been hospitalized by mistake and who would function perfectly normally once released.

By the mid-1960s, then, it had become politically correct to denigrate psychiatric hospitals and assert that the patients would do just fine if moved to community settings. The ethos was similar to the British film King of Hearts, in which the inmates of the asylum are released en masse, take over the town, and live happily ever after. Civil libertarians and traditional liberals promoted this idea; fiscal conservatives, who saw deinstitutionalization as a way to save state funds by shutting down some of the hospitals, went along too.

In retrospect, it is easy to see the deeply flawed assumptions undergirding deinstitutionalization. One such assumption was that nothing much was wrong with patients in the psychiatric hospitals. At that time, the Freudian psychoanalytic theory prominent in American psychiatry held that bad mothering and social stresses by themselves could lead to disorders such as schizophrenia. Research in the past decade has shown this assumption to be false: studies using such techniques as magnetic resonance imaging (MRI) and positron emission tomography (PET) scans have proved that schizophrenia and manic-depressive illness are physical disorders of the brain in exactly the same way as Parkinson's disease or multiple sclerosis. Patients with such illnesses need medications to control their symptoms; without the medications, their symptoms usually get worse.

The second flawed assumption was that mentally ill individuals discharged from the psychiatric hospitals would voluntarily seek psychiatric treatment if they needed it. Psychiatric professionals in the 1960s and 1970s pointed to the newly funded Community Mental Health Centers (CMHCs) as places where the mentally ill could go to get continuing treatment.

In reality, however, most CMHCs had no interest in treating the patients being released from public psychiatric hospitals. Their staff psychiatrists aspired to do counseling and psychotherapy for individuals with more ordinary problems, just like psychiatrists in the private sector. One federally funded CMHC in Wisconsin even made individuals with schizophrenia use a separate clinic door and waiting room so that they would not offend the "worried well" going there for psychotherapy appointments.

More seriously, however, approximately half the patients discharged from state psychiatric hospitals did not seek treatment once out of the hospital because they did not believe themselves to be ill. In psychiatric terms, they lacked insight into their condition. Recent studies have confirmed that approximately half of all individuals diagnosed with schizophrenia and manic-depressive disorder have significantly impaired insight into their condition. They believe that the Russian KGB really is following them, or that they really did write songs for the Beatles and are owed millions of dollars in royalties. The fact that they lack such self-awareness is not surprising, since as we now know, these illnesses often afflict the same parts of the brain that we use to think about ourselves and our needs.

In practical terms, this means that approximately half of those who suffer from schizophrenia and manic-depressive disorder will not seek treatment voluntarily no matter how attractive the treatment facilities might be. And since lawyers working for the American Civil Liberties Union and the Bazelon Center in Washington, D.C., have changed state laws to make it exceedingly difficult to treat the mentally ill involuntarily, roughly half of these individuals are untreated at any given time. They constitute most of the mentally ill population who are homeless or in jail, and who commit acts of violence.

Finally, advocates of deinstitutionalization wrongly assumed that the states would continue to take responsibility for treating mentally ill individuals after discharging them from public psychiatric hospitals. This was a reasonable assumption, since states had had virtually complete responsibility for treating these individuals since the early nineteenth century. For example, in 1963, when deinstitutionalization was getting under way, state and local government contributed 98 percent of the total cost for support and services for persons with mental illnesses, and the federal government contributed just 2 percent.

All that changed in the 1960s, when Washington made mentally ill individuals who had been discharged from state psychiatric hospitals eligible for a variety of federal programs. Funds from Medicaid, Medicare, Supplemental Security Income (SSI), the Disability Insurance Trust fund (SSDI), food stamps, and housing programs all became available to those with serious mental illnesses. States quickly learned how to shift the costs and the responsibility for providing public psychiatric services to the federal government, so that by 1994 the state and local share of total costs had decreased from 98 to 38 percent, while the federal share had increased from 2 to 62 percent. As the discharged psychiatric patients increasingly fell between the cracks in the treatment system, the states increasingly disavowed responsibility for them.

Given the vast scale of deinstitutionalization, one might have thought that both the federal and state governments would have been interested in evaluating the process and assessing the outcome, especially since this social experiment had been launched with virtually no empirical base. Incredibly, this did not happen. As late as 1981, when deinstitutionalization had been under way for over 15 years, an academic review of research on the subject found only five studies concerned with outcomes, three of which it deemed methodologically flawed. During these same years, the National Institute of Mental Health evaluated the Community Mental Health Centers program and discovered that patients being released from state psychiatric hospitals were not, with only occasional exceptions, receiving aftercare from the CMHCs. In other words, during the mass exodus of patients from psychiatric hospitals, nobody bothered to ask what was happening to them. It is doubtful whether, in the history of American medicine and social services, a policy of this magnitude has ever been implemented with as little evaluation or follow-up.

Given the debacle of deinstitutionalization, what can be done to correct the situation? Two changes would be most helpful.

First, responsibility for mental illness services should be fixed at the state and local levels, and the states should then be held accountable. Human services is not a function that the federal government carries out well. Let states with populations of under 10 million take ultimate responsibility for providing public services for individuals with psychiatric disorders. In states with populations of over 10 million, including New York, counties should take ultimate responsibility, as has happened in California.

States and counties cannot assume responsibility, however, without the necessary fiscal resources. Accordingly, federal funds now being used for mental illness services, including federal Medicaid and Medicare funds, should be given to the states in block grants. In most states more money than is presently being spent is not needed; what is needed is flexibility to spend the funds in the most efficient manner to provide good services. Turning federal funds into block grants would make it possible for patient needs, rather than reimbursement regulations (as is unfortunately the case today), to be the main force shaping the mental illness treatment system.

How would mental illness services change? States would doubtless discover that closing down all state hospital beds is ultimately not cost-effective and that a small percentage of seriously mentally ill persons need long-term inpatient care. The states would also discover that it costs less to maintain the mentally ill on medication—by making it a condition of their remaining at liberty—than to have them readmitted to the hospital every few months.

With responsibility should come accountability. State mental illness services should undergo an annual or biennial evaluation, funded by the federal government but carried out by a private contractor (RAND, say, or the University of Michigan's Institute for Social Research). Federal agencies cannot carry out such evaluations effectively because they are too susceptible to congressional pressure, White House mandates, and political correctness. Measures would include quality-of-life ratings (both by the patients and by their families), program evaluations, and data on the number of patients who are employed, homeless, or in jail. The results of the state evaluation would partially determine the size of the next federal block grant, with states being penalized for doing poorly and rewarded for doing well.

A second key reform that would improve today's bad situation is more controversial: the mental health system must provide for the occasional involuntary treatment of seriously mentally ill individuals. Of the approximately half of all seriously mentally ill persons who have impaired insight into their own condition, a significant fraction will need such treatment once initial attempts to persuade them to accept help have failed.

Involuntary treatment requires a court order for commitment, either as an inpatient or outpatient, or for guardianship. Outpatient commitment, commonly used in 12 states, requires the person to continue taking medication if he wishes to stay out of the hospital and has proved especially effective in reducing rehospitalizations. Thanks no doubt to the power of its civil-liberties lobby, New York has been the only state with a statute specifically prohibiting the use of outpatient commitment, although legislators modified this law in 1995 to allow a pilot program for outpatient commitment at New York City's Bellevue Hospital.

The crux of any commitment law is the conditions it sets for involuntary commitment to be legal. In many states, posing a danger to oneself or others is the only grounds for commitment, and courts often interpret this provision very strictly. In order to treat seriously mentally ill individuals before they have proved dangerous, it is necessary to include "need for treatment" or "substantial deterioration" as criteria for involuntary treatment, and several states have moved in this direction. There should, of course, be provisions for court hearings and public defenders to ensure that involuntary treatment is not abused, as it was in the Soviet Union. But as columnist (and psychiatrist) Charles Krauthammer has noted, the use of dangerousness alone as the criterion for involuntary commitment "is not just unfeeling, it is uncivilized. The standard should not be dangerousness but helplessness. Society has an obligation to save people from degradation, not just death."

A major danger in thinking about the disaster of deinstitutionalization is the temptation to accept it. An entire generation of young adults has grown up seeing homeless mentally ill individuals living on the streets and in the parks. From their perspective, why shouldn't these people always live there? To them, the Joyce Browns and Larry Hogues of the world are simply part of the urban landscape, along with broken-down cars at the curbs and refuse under the bridges.

It is important for those of us who are older to speak out. We remember when homelessness was rare, confined to alcoholic hoboes. We must not accept the debacle of deinstitutionalization and the inevitability of its consequences. We made this problem, and we can correct it.


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next