In the 1980s the experiences reported by overwhelmed service providers were confirmed by epidemiological studies indicating the emergence of a wave of homeless people with psychiatric disorders. By the late 1980s, reliable studies describing psychiatric illness showed reasonably consistently that one-third to one-half of homeless people had severe psychiatric disorders, such as major mood disorder (19 percent to 30 percent) and schizophrenia (11 percent to 17 percent) (3,4,5,6). Investigators also reported rates of drug abuse among this population from 20 percent to 30 percent and rates of alcohol abuse from 57 percent to 63 percent (3,4,7). The phenomenon of the dual diagnosis of chemical dependence and mental illness in the homeless population began to receive attention (8).
People with psychiatric disorders had been clearly swept into the tide of homelessness brought on by macroeconomic shifts, including the loss of unskilled industrial employment, the rise of a service economy, and especially an emerging shortage of low-income housing. Given the coincidence of homelessness with the decline in the numbers of patients in psychiatric hospitals, some observers also viewed the homelessness epidemic among mentally ill persons as an unforeseen by-product of deinstitutionalization (9,10). For example, in 1970, there were 413,066 beds in state and county mental hospitals in the United States. By 1988, this number had decreased to 119,033; by 1998, the decline had continued to 63,526 beds (11).
Public attention to single adults who had severe psychiatric disorders, and to a somewhat lesser degree chemical dependency, helped focus the psychiatric profession on the clinical care of these populations in particular. Psychiatric interventions began primarily with volunteer initiatives that evolved into a relatively small culture of paid psychiatrists who were experts in homelessness—generally working in special programs in community-based organizations rather than hospital systems. In part, this situation reflected the traditional health system's hesitancy to adapt to the needs of an alienated and poorly insured patient population that is often viewed as presenting medicolegal risks. Bachrach (12) observed that the mentally ill homeless were "generally seen as dirty, smelly, unresponsive to traditional treatment modes, and disruptive in traditional treatment settings."
Nevertheless, as the national crisis in homelessness evolved into an endemic social problem, a parallel consensus emerged about the kinds of services needed for engaging mentally ill homeless people in treatment (13). Psychiatrists developed some of the programming that contributed to this consensus, especially with respect to acute and emergency treatment. An example of this was Project HELP in New York City, which pioneered street outreach blended with involuntary emergency intervention (14). Project HELP was later functionally linked to a program that provided acute inpatient stabilization and extended transitional care in the community through a unique municipal-state collaboration between Bellevue Hospital and Creedmore Psychiatric Center (15,16).
Important service principles were subsequently articulated, especially in relation to psychiatry, as evidenced by the recommendations of the APA task force on the homeless mentally ill for outreach, assertive treatment, rehabilitation, and housing (17).
Through these efforts, and because of the hard work of committed individual psychiatrists, a range of professional roles began to evolve for psychiatry in relation to homelessness. These roles encompass clinical, administrative, academic, and advocacy functions. Below we describe these roles, the degree to which each has developed, and the current challenges involved in carrying them out.
We continue with a discussion of how homelessness in America is undergoing flux and how psychiatry's roles must further adapt to general epidemiological trends and the service delivery implicit in these trends.
Some consensus has emerged on how to help mentally ill homeless people reintegrate into the social mainstream. It is useful to consider a transitional process that includes stages from homelessness through permanent housing (18). In one APA task force report, these rehabilitative stages are engagement, intensive care, and ongoing rehabilitation (19). They begin with engagement through outreach on the street or in shelters, where homeless people are introduced to a range of social and medical services—optimally ending with housing, work, and the possibility of reestablishing lost social ties. Throughout the three stages exists a continuum of services and interventions that have different levels of technical development. The details of these stages, a topic beyond the scope of this article, are described elsewhere (20), but three issues are noteworthy. These issues have developed over the past ten years and continue to present challenges to clinical psychiatrists as a consensus about services continues to evolve.
The first issue is the relatively nascent status of knowledge concerning what contributes to housing stability. The best of recent research is exemplified by a naturalistic study of 2,937 people with mental illness who obtained housing placements in New York City (20). In high-intensity housing—generally, housing with ample staffing, medication dispensing, money management, curfews, and on-site meals—it was found that noncongregate living arrangements and medication management were associated with longer housing tenure, but in low-intensity housing different associations were found. Clearly, this type of research is only a beginning, and researchers have yet to describe methods for creating a spectrum of housing types based on clusters of personal needs (21). Little information is available that analyzes individual models or programs (22,23) or that systematically examines clinical issues like medication adherence, concurrent treatment of substance abuse (24,25), medical comorbidity, and family involvement. In fact, novel programs have been devised that attempt to address housing as a rehabilitative technology—a type of intervention that is in a relatively early phase of development. Critical time intervention focuses on extended transition and follow-up by shelter workers who assist clients as they attempt to adapt to housing programs (26). Another innovative response, the Pathways to Housing program, bypasses established housing programs altogether (23). It employs the assertive community treatment model to aggressively maintain its clients at its own scatter-site housing, seeing them though a process culminating in housing stability.
In our view, clinical psychiatrists working with this population should actively involve themselves in decisions concerning patients' housing placement. This activity represents a leadership function for clinical psychiatrists within community rehabilitation teams, regardless of whether the psychiatrists are consultants or administratively in charge. A still broader objective is that clinical psychiatrists use their expertise to identify the personal characteristics that will help them understand housing needs and take leadership in fostering program evaluation and research. Community-based programs typically do not have the capacity for research, but services research can be initiated by persuading program administrators to participate, developing academic liaisons, and collaborating with other disciplines that have affinity with research. This issue is particularly timely because of a rapidly advancing emphasis on evidence-based mental health programming and funding that is increasingly attuned to documented outcomes.
The second issue concerns vocational training, an area of rehabilitation particularly relevant to homelessness and one increasingly scrutinized through outcomes research. Once homeless people have found satisfying housing, the rehabilitative focus usually shifts from symptom management to helping them develop a new set of social roles. A vocational identity is important to this development. As the struggle with daily survival recedes, people often begin to seek financial independence as a source of pride. Supported employment has been the most fully researched strategy (27), but other models, such as transitional employment programs in clubhouses, have a rich service history too. It is likely that different vocational models suit different individuals. For homeless people, goal setting related to work may actually begin before housing placement, and initial attempts have been made to adapt supportive employment, for example, to homeless populations (28). Because of the significance of work to role redefinition among many mentally ill homeless people, clinical psychiatrists must become conversant with vocational training.
The third issue concerns mentally ill homeless people in the criminal justice system. The U.S. Department of Justice estimated in 1999 that 16.3 percent of detainees in jail reported either having an emotional condition or having spent a night in a mental hospital or program (29). Although these are highly inclusive criteria, they illustrate how the mental health and criminal justice systems are sharing a common responsibility. The same study showed that among jail detainees with mental illness, 30 percent were homeless when they entered custody, compared with 17 percent of those without mental illness.
It has been recognized since the late 1980s that a large number of homeless people with mental illness have contact with the criminal justice system (30,31) and occupy a closed transinstitutional circuit between shelters, hospitals, and incarceration (32). Although the overall causes of this situation may include a decrease in housing and jobs, it has been proposed that there are issues unique to those with mental illness, including access to care, society's ambivalence about whether to treat or to punish, and the effects of deinstitutionalization and restrictive commitment laws (33). What is clear is that jails and prisons are common institutional links in the care of mentally ill homeless people.
This phenomenon has forced the criminal justice and mental health systems to explore ways to work effectively together—from the police to the courts to jails and prisons (34). Among the creative solutions to emerge are projects that divert people from incarceration (35,36) and programs whose function is to reintegrate mentally ill parolees into the community (37). Psychiatry's challenge is to lead efforts in training members of the law enforcement and other criminal justice professions to work effectively with mentally ill people. Psychiatrists must also help jail and prison mental health services move their focus away from custodial care toward integrating inmates into community-based psychiatric rehabilitative services, thus helping to break the transinstitutional circuit. Conversely, as psychiatrists increasingly relate to the criminal justice system, they will more patently exercise social control. In doing so, they must understand how to work within the criminal justice system while remaining clear about where their role as advocates emerges, whether on an individual or a systemwide level.
In the 1970s, psychiatrists began to lose their initial administrative dominance in community mental health centers while maintaining it in traditional systems like hospitals (38). Owing in large part to this change, nonmedical mental health professionals sought and were able to fill an important niche in leadership in community mental health centers and other community-based organizations. Because of a need to respond creatively to the particularly marginalized clinical population of mentally ill homeless persons, some of these new leaders and their organizations developed nontraditional approaches to services. In addition, community-based organizations usually have more administrative flexibility and in many cases a broader definition of what constitutes service delivery than do traditional settings. As a result, many organizations that work with homeless populations have developed considerable breadth, encompassing assertive outreach and case management, shelter care, housing, and vocational rehabilitation.
Service delivery methods at community-based organizations often embrace rehabilitation as a cardinal precept (39). Rehabilitation, despite its identification as important in the psychiatric care of mentally ill homeless people (19), is an approach that has still to gain acceptance in psychiatry (40)—which is at least one reason why community-based organizations tend to be skeptical of traditional mental health care and the leadership positions that psychiatrists have usually held within it.
Nevertheless, over the past decade, the role of medical director has gradually emerged in these community-based organizations (41), representing an adaptation of psychiatric administration to include nontraditional models and the acceptance by many community-based organizations of a meaningful medical contribution. Reasons for this change may include a greater emphasis on teamwork in psychiatric education, more understanding by the staffs of community-based organizations of the importance of biological treatment, and the formation of community-hospital care systems through public-sector managed care—thus requiring that community-based organizations add to their staffs physicians who can authorize "medical necessity." In a study by Ranz and colleagues (42), it was found that, among community psychiatrists working in community-based organizations, job satisfaction was significantly higher among medical directors than among staff psychiatrists—implying that medical directorship is an area of potential development for psychiatrists in community settings, including homelessness services programs (43).
The shift in the relationship between psychiatry and community-based systems has introduced challenges. The most obvious one is the relationship between medical directors and nonmedical administrators. Medical directors work in partnership with, or under the aegis of, nonmedical program or agency directors. How the relationships are managed is of critical concern, and it requires flexibility and collaborative skills similar to those that psychiatric clinicians use in teamwork. For example, in the area of clinical accountability of nonmedical staff, the medical director needs to assert a commitment to high-quality care, often without having direct line authority.
Psychiatry's role in nonclinical affairs is a related challenge. Many community-based organizations operate programs whose focus is less explicitly clinical, such as vocational training for people recovering from substance abuse. For this kind of program, medical directors or chief clinical officers must be able to see beyond being physicians in the narrower sense; they must understand how psychiatry can support and inform a social agency's broader agenda. This brand of leadership offers potential for psychiatrists to extend their role to that of executive director. Such a development within the culture of community-based organizations would signify further evolution for psychiatric leadership capabilities. How education can address this is discussed in the next section.
Faculty who came of age during the phases of the problem of homelessness lead today's teaching experiences. Some faculty have begun to define an emerging clinical technology for work with mentally ill homeless populations (19,44,45,46,47,48). These teachers understand that such work gives trainees an opportunity to learn a form of psychiatric intensive care in which they exercise diagnostic and management skills in treating complex and very distressed patients. Through these experiences, trainees also work at the frontiers of epidemics like tuberculosis, HIV illness, and hepatitis C.
Training physicians to treat mentally ill homeless people starts with teaching trainees a theoretical framework encompassing the social and economic roots of homelessness and how community psychiatry fits within applied public mental health. Successful clinical experiences last long enough to allow hard-won relationships to develop between patient and clinician, showcase successful mental health programs, and involve two support levels: a mentoring experience through expert supervision and trenches-level collegiality from both medical and nonmedical on-site staff. In this model, supervisors teach cultural sensitivity, work with countertransference responses, and provide special clinical techniques.
There are barriers to this type of training in residency programs and medical schools. A recent survey found that 51 percent of American psychiatry residencies offered clinical experiences dedicated to homeless populations but that only 5.6 percent of surveyed programs rotated all residents through such experiences (unpublished data, McQuistion HL, Ranz JM, Gillig PM, 2002). As educational programs attempt to establish training experiences beyond the walls of their host institutions, limitations to community-based training may be encountered, such as fiscal constraints and a lack of faculty availability. Although training in the care of homeless people with psychiatric illness has expanded over the past two decades, there may not be enough faculty in relevant community settings to supervise residents and medical students. Also, in our experience, academic programs rarely make vigorous efforts to recruit community psychiatrists into training roles.
Moreover, if the cost of clinical training in homelessness services programs cannot be covered, training experiences will not thrive. Until academic programs respond as they should, creative funding mechanisms need to be explored. Community agencies can finance residency or faculty positions or cooperatively engage with academic departments to secure training grants—a necessary step toward introducing academia into the community. In one such initiative, New York City's Project for Psychiatric Outreach to the Homeless acquired funds to encourage several residency programs to place trainees in agencies where homeless people are served (49).
Academic education has an additional role in introducing psychiatric leadership into community-based organizations. Medical student and psychiatric residency training primarily serves clinical competence, and attention to systems of care is relatively unusual. To teach residents that there are ways to manage and change systems enhances their professional identity. Therefore, residents should be exposed to leadership and management principles: how human services and health care intersect, how care systems are organized and funded, how quality is measured, and how to use entrepreneurship and social enterprise.
When programs methodically teach residents these principles and enable them to train in community settings, the results can be remarkable. The public psychiatry training program at the University of Oregon is perhaps the best established of such programs, combining meaningful community rotations and well-refined, systems-oriented didactics. Over 29 years, 43 percent of the trainees graduating from the residency that hosts the training program had at least part-time employment in community mental health settings (50).
Similarly, advanced degrees in public administration, business administration, and public health prepare psychiatrists for administration, as can postresidency community psychiatry fellowships such as Columbia University's fellowship in public psychiatry, which focuses on leadership and management. A recent survey of its graduates showed that 58 percent had positions as medical director (51). For a psychiatrist who takes on a leadership role in a community-based organization, this kind of further training in combination with clinical expertise confers well-earned credibility.
People who are mentally ill and homeless are among the least politically empowered groups in America—experiencing in addition the stigma of being indigent within a materially competitive society. They often also suffer from other stigmatizing conditions, like minority status, HIV illness, criminal history, and substance abuse (32,52,53).
To advance recovery and help mentally ill homeless people enter the social mainstream, advocacy must be integrated into all psychiatric roles. At a basic level, individual psychiatrists' work with homeless people constitutes advocacy by drawing professional attention to the issue. At the program level, psychiatric administrators must aggressively form service alliances, combating fragmentation of care. Psychiatrists working with mentally ill homeless people must also influence or direct the organizations they work for to create and practice a corporate function in advocacy, moving beyond service. Although some organizations may view this stance as adversarial and complicating, raising worries about jeopardizing funding, it is important that organizations articulate the social issues that have institutionalized homelessness in American culture. Because they are front-line programs, their credibility can affect public awareness, and their skilled advocacy will ultimately improve resources for funding.
In some localities, litigation has been a critical mover of public policy on homelessness. Psychiatry offers unique technical expertise to legal advocacy, and this role is particularly potent in facilitating social change. In this area, psychiatrists have given expert testimony in important class action suits (54,55,56). Organized psychiatry, especially, can be instrumental in this regard, and professional organizations have also issued expert amicus curiae briefs in class actions (57).
All mental health disciplines, including psychiatric nursing, social work, and psychology, have historically contributed much to advocacy in this field on both individual and organizational levels. The production of influential reports is an area in which organized psychiatry has been at the forefront. As noted above, APA has published two highly influential reports drawing professional attention to the mentally ill homeless population (1,2) and also published a 1991 report on homeless families and children (58). These publications serve a variety of purposes. They document current knowledge, are policy reference points, make declarations of professional commitment, and provide tools for networking with other advocates and concerned professionals. These reports thus enable psychiatrists to provide leadership in the public discourse on homelessness.
These efforts notwithstanding, it is unclear how focused and, therefore, how effective organized psychiatry has been in advancing its opportunities for leadership in advocacy. Professional psychiatric organizations, such as APA, could be even more effective if their advocacy were systematically coordinated with similar professional organizations that could take or have already taken similar positions on issues of poverty, homelessness, and mental health. As a case in point, the American Psychological Association (59) has also issued position statements addressing the problem of homelessness in America, but there has been little if any coordination between the two organizations.
As it has done with other groups on issues like mental health parity (60), the profession could forge alliances with homelessness advocates, including those concentrating on criminal justice, health care, consumer rights, family, and housing. Through its expertise in research and clinical care, psychiatry can occupy the important niche of taking the lead in advocating for homeless people with identified psychiatric disorder—a term that has traditionally meant serious mental illness. However, as the demographics of homelessness change, psychiatry will need to identify the mental health needs of other clinical populations who are homeless, enabling the profession to become expert in the advocacy needs of these populations.
During the 1990s, studies emerged indicating that homelessness is far more prevalent than had been understood, even by some advocates. Independently, two well-executed studies (61,62) estimated that approximately 3 percent of Americans experience homelessness in a five-year period. Both investigations also highlighted a social mobility in homelessness, finding that most homelessness episodes are brief. Although these studies found that relatively few people have recurrent episodes, emerging data indicate that over a ten-year period, recidivism can be as high as 28 percent (unpublished data, New York City Department of Homeless Services, 2002).
Moreover, we are now witnessing an alarming resurgence of homelessness. Homelessness among single adults has remained at an unacceptably high level and has risen dramatically among families and children. This new wave of homelessness evokes clinical and service questions that are still unanswered. Thus the need is as urgent as ever to understand better the causes of homelessness and solutions to it.
Evidence of increasing homelessness among families comes from several sources. Prevalence data from New York City show a significant rise in homelessness in families, especially over the period 2000 through 2002. On an average day in 2002, there were 24,517 family members in New York City's municipal shelter system, a rise of 46.1 percent from 16,778 average daily family members in 2000. Simultaneously, the number of single adults in the system rose by 13.4 percent, from 6,934 to 7,863. During the prior two-year period, average daily family and single adult populations rose by 15.8 percent and .8 percent, respectively (unpublished data, New York City Department of Homeless Services, July 2002). Despite methodological limitations in compiling information collected by individual communities, the U.S. Conference of Mayors (63), which culls its findings from other cities, has corroborated these New York data. Its survey shows a 14 percent average national yearly rise in homelessness from 1998 through 2002, as opposed to 8.4 percent in the previous five-year period. This document also shows a greater rise in homelessness among families with children than among single adults.
The special needs of homeless families have been apparent for at least the last decade (58,64). Metraux and Culhane (65) offered glimpses into repeated episodes of homelessness among women with families, demonstrating that the factors of domestic violence, new motherhood, and children in formal or informal foster care were associated with significant risk of repeat shelter stays in New York. Family instability, low social support, and trauma in adulthood relate to known predictive childhood experiences, linking out-of-home placement and running away to adult homelessness (66,67,68). Herman and colleagues (69) showed that a lack of primary caretaker contact in childhood increased the likelihood of homelessness. Study participants who reported lack of care were also likely to experience sexual and especially physical abuse. Experiencing all these factors afforded a risk of eventual homelessness 26 times higher than the risk for those who did not experience them. A pattern of family instability and despair is underscored by evidence that significantly more children who are in foster care have biological parents who are homeless (70). For many, then, the rough and lonely life of the street and shelter is part of a cycle of continued trauma and isolation.
These are important threads of emerging knowledge about the mental health vulnerabilities of a growing cadre of homeless people. Therefore, psychiatry, while it continues to offer crucial expertise in caring for those with severe psychiatric disorders, must also turn its energies toward the majority of homeless people who have more subtle, yet complex and debilitating, problems. An important challenge in meeting this goal is to support research that seeks to discover the combination of psychological and social factors common to homelessness, discerning whether there are differences in psychiatric vulnerabilities between homeless and domiciled poor persons. Some investigations indicate that substance abuse may serve as a marker (71,72,73) through its corrosive effects on adaptive social ties and psychological functioning. A greater understanding of the vulnerabilities of these individuals and their families will illuminate mental health initiatives and will consequently inform community-level social interventions that build social and economic capital, preventing homelessness and offering opportunities for homeless persons to rediscover constructive social roles.
All mental health professions have responded to the emergence of homelessness as a persistent feature on the American scene. Mainstream psychiatry's response has been relatively slow, but the profession has played a role as epidemiological and services research has begun to articulate the causes of homelessness and to inform effective rehabilitative interventions. In this context, we have shown that psychiatry has developed identifiable roles, especially with respect to working with homeless people with serious mental illness.
These roles are still evolving, and they deserve reevaluation and renewal now, as well as periodic renewal as times change. Psychiatry's roles will continue to evolve if the profession can keep pace with epidemiological and social trends that affect the character of homelessness, especially with respect to addressing the breadth of psychiatric disorders among homeless people. In addition to its past contributions, psychiatry has social legitimacy and, at its best, unparalleled biopsychosocial expertise. In our opinion, the psychiatric profession is therefore uniquely positioned to help shape the social discourse on homelessness by continuing to focus on meeting the challenges of the roles it has developed.
A key to keeping this focus is prioritizing advocacy across all the roles of psychiatry. In doing so, the profession can deploy its strengths in the natural and social sciences and practice its commitment to patients' rights and well-being. Thus psychiatry will not only achieve credibility and effectiveness for individual patients but also enhance its leadership in the broader quest for solutions to end homelessness.