U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) Committee on Health Care for Homeless People. Homelessness, Health, and Human Needs. Washington (DC): National Academies Press (US); 1988.

Cover of Homelessness, Health, and Human Needs

Homelessness, Health, and Human Needs.

Show details

5Health Care Services for Homeless People


To the extent that homeless people have been able to obtain needed health care services, they have relied on emergency rooms, clinics, hospitals, and other facilities that serve the poor. Indigent people (with or without a home) experience many obstacles in obtaining health care. For homeless people there are additional barriers. Recognition of the special health care needs of homeless people has encouraged the development of special services for them. In observing and describing these health care and health care-related services, one must be mindful of the heterogeneous nature of the homeless population, as well as the structure of the communities in which such services have developed. Regardless of differences among homeless people or regional variations in services, however, homeless people are more susceptible to certain diseases, have greater difficulty getting health care, and are harder to treat than other people, all because they lack a home. Similarly, attempts to provide health and mental health care services, regardless of variations in such areas as history, funding levels, and nature of support, also have certain common elements. They arose in response to a crisis rather than developing as part of a well thought out plan. They generally brought services to homeless people rather than waiting for them to come in; increasingly, they rely on public funding because the problem has grown beyond a level that the private sector can support.

The purpose of this chapter is to describe programs that seek to bring general health and mental health care services to homeless people. The information presented in this chapter is largely based on the 11 site visits made by members of the study committee and its staff. Although the sites are not representative of the entire universe of programs for the homeless, they were selected to include the broadest range of programs possible and to be geographically dispersed throughout the country.

In studying health care and related services for homeless people, the committee sought to examine a broad range of services developed over a period of time, rather than to focus only on specialized services or services that have been developed recently. However, what the committee observed were discrete services and programs. At no time did the committee encounter anything that could be appropriately called a ''system" of services.

Before describing how these various programs bring general health care and mental health care to the homeless, we must address two major issues: (1) what makes serving the homeless, in contrast to the indigent in general, more difficult?; and (2) based upon the literature and the site visits, what elements enhance a program's ability to provide such services to this population? In Chapter 3, we discussed those aspects of treatment that are especially difficult to implement when the patient is homeless. However, one must also look at the people who are homeless. William Breakey (in press) has identified characteristics of homeless people that affect the provision of treatment and the planning of health care services:

Daily Activities—Some homeless people live under circumstances that pose particular problems for developing a treatment plan. For many, it may be difficult to keep a supply of medication while living on the street. For an alcoholic trying to stay sober, a homeless existence may present too many opportunities for drinking. Some former patients complain that neuroleptic medications, prescribed for a schizophrenic illness, may make them too drowsy and interfere with their alertness against the dangers on the streets.

Multiplicity of Needs—In addition to physical and mental health problems and difficulties with such things as housing and income maintenance, homeless people often also suffer from drug or alcohol abuse. Any health care program for homeless adults should expect that 25 to 40 percent of patients will suffer from serious alcohol or drug abuse problems (Fischer and Breakey, 1986).

Disaffiliation—Although many homeless people establish individual support networks outside a family structure, some homeless people typically lack those networks that enable most people to sustain themselves in society. Such isolation often causes (and sometimes is caused by) a limited capacity to establish supportive relationships with other people. Difficulties in establishing and maintaining relationships can militate against the development of cooperation with health care providers and may be an important factor in explaining what is often inaccurately described as a "lack of motivation."

Distrust—In addition to their distrust of authority, many homeless people are disenchanted with health and mental health care providers. Some have had bad experiences with medications, hospitals, doctors, and other human service professionals and are leery of further involvement.

Except for anecdotal information and obvious indicators of utilization, it is not possible to assess the effectiveness of health care delivery systems for homeless people. There are no adequate data from which such assessments can be made. However, in its review of various programs for health and mental health care services for homeless people, the committee found that four common elements enhanced a program's ability to provide services to this population:

Communication—Those people and agencies involved in the effort to address the health care problems of homeless people interact regularly and frequently.

Coordination—Even if only in a most rudimentary form, there is some way in which clients can be linked with a wide range of existing services (i.e., health and mental health care, housing, social services, entitlements, etc.) by providers, rather than being forced to seek services without assistance.

Targeted Approach—Programs are aggressive in seeking the homeless, rather than passive in waiting for them to appear. This may be reflected by locating a program in a skid row area. Other programs provide outreach and seek out homeless individuals on the streets.

Internal and External Resources—These constitute the range of resources that a program requires to carry out its function adequately, no matter how limited that function might be. Internal resources include reasonable funding and paid employees, in addition to the utilization of volunteers and donated goods and facilities. External resources include both the network of essential services described above and the ability to access that network.

The Health Care for the Homeless projects, funded jointly by the Robert Wood Johnson Foundation and the Pew Memorial Trust, are considered by many to have been the single most effective network of health care services developed for homeless people in the 1980s. They are also generally viewed as providing a major impetus for Title VI (health care) of the recently passed Stewart B. McKinney Homeless Assistance Act of 1987 (P.L. 100-77). The first nationwide program to address the health care problems of the homeless, the projects' creation serves as a benchmark. Therefore, this chapter is arranged from the perspective of that unique role. The following sections of this chapter describe: (1) programs in existence prior to the Johnson-Pew projects; 1 (2) the Johnson-Pew program itself; and (3) other programs that came into existence at roughly the same time (1984-1987) as the Johnson-Pew projects. The description of the third group is further subdivided, based upon the targeted populations. The final section of this chapter discusses various programmatic, administrative, and clinical issues identified throughout the course of the committee's observation of these service delivery models.

Pre-Johnson-Pew Models

Several program models were developed to provide health care services to homeless people before the mid-1980s. The conclusion that they are effective models of service delivery can be drawn from their reported experiences and the fact that the major features of such models appear repeatedly in later programs (especially the 19 Johnson-Pew projects).

Shelter-Based Clinics

Shelter-based clinics provide the types of services most frequently found throughout the country. Recognizing a need to bring services to where homeless people can be found, those involved with shelters or health care have developed on-site clinics at shelter locations.

Rescue Missions

The committee visited volunteer clinics located at rescue missions in Kansas City, Los Angeles, Nashville, and San Diego. These rescue missions are coordinated on the national level by the International Union of Gospel Missions, but there is an even greater strength of coordination locally. Having served the homeless for extended periods, they are known to the community and have substantial access to existing networks of, for example, health care services, housing, and social services. The clinics tend to be staffed by volunteer doctors and nurses and rely heavily on private donations, both of cash and pharmaceutical and medical supplies (although some have begun to accept limited financial support from local governments). However, because of the religious aspects of the organizations that operate these clinics, not every homeless person is willing to go to them.

Nonsectarian Programs

Nonsectarian programs, such as the clinic at the Pine Street Inn in Boston, operate similarly to the religious rescue missions. They have developed strong sources of financial support, frequently from among local businesses, charitable organizations, and foundations. In the absence of any national coordinating or controlling body, they tend to reflect the characteristics and needs of the city in which they are located.

Both the rescue missions and the nonsectarian programs face certain common problems: limited hours (many shelters are closed during the day), dependence on volunteers, limited access to some of the less common medications, limited specialty and ancillary services (e.g., podiatry and dental care), lack of an ability to perform systematic screening, and difficulty in obtaining both liability insurance and medical malpractice insurance (especially critical when volunteers are retired physicians who do not have their own malpractice insurance). Both the rescue missions and the nonsectarian programs are, however, major sources of private, not-for-profit, and non-tax-supported health care for homeless people.

The Public-Private Programs

Public-private programs share some of the attributes of all volunteer clinics, but they have often resolved some of the problems cited above. One of the oldest examples is the St. Vincent's Hospital and Medical Center Single Room Occupancy (SRO) and Shelter Program in New York City.2 The initial program developed from an intern's concerns over the large number of people who arrived by ambulance from one SRO hotel. Outreach programs were designed to provide health and social services on-site at SRO hotels and municipal shelters. With some variance according to the site at which services are provided, an interdisciplinary team of a physician, a nurse, and a social worker established on-site medical clinics. In recent years, partial funding for the program has been received from the New York City Human Resources Administration, that city's department of social services. In addition to the benefits of on-site programming, the clinics and the Department of Community Services at the hospital closely coordinate their efforts. Homeless people referred to the hospital for specialized services are often treated by the same individuals whom they saw at the on-site clinic, improving the continuity of care and increasing cooperation with the care-giver.

Health Care Services in Day Programs

Day programs, which are similar to the shelter-based clinics identified above, provide services where homeless people can be found, but they differ from shelter-based clinics in that the sites are independent of residential programs. One good example is St. Francis House in Boston, which has been described by its staff as "a shopping mall of services to the homeless." Various mental health and vocational guidance services are provided to homeless people in a single building located in what was once known as the "combat zone"3 of Boston. Included in these services is a health clinic for homeless people that is staffed by volunteers and paid employees.

A similar program, also in Boston, is the Cardinal Medeiros Day Center operated by the Kit Clarke Senior House. Located in a church in downtown Boston, this is a day program exclusively for elderly homeless people. Among its services is a food van that stops where the elderly homeless are known to congregate. A registered nurse who is part of the van team performs basic health assessments and referrals for anyone willing to accept this service. A second nurse, stationed at the Medeiros Center, provides more extensive services. The two nurses alternate between the van and the center, so they are familiar with both programs and are readily identified by the homeless people themselves. While the nurse reported to the site visit team that there is little opportunity to perform other than the most basic visual assessment of a homeless person's health status from the van, she indicated that the true value of the program came from gaining the confidence of homeless people and then referring them to the Medeiros Center at a time when she could perform a more detailed assessment. The fact that they knew her enabled them to overcome any fear that might have prevented them from seeking health care.

A third program of this type is So Others Might Eat, known as SOME, a day program in Washington, D.C., whose primary purpose is to provide breakfast and lunch to homeless people. Since 1982, SOME has been the site for a medical clinic operated by the Columbia Road Physician Group, a group practice composed of four physicians committed to serving homeless and indigent people and providing on-site social services and substance abuse counseling. It has also been the site for a dental clinic operated by the Georgetown University Dental School.4

Free-Standing Clinics

In 1979 a somewhat different model for the delivery of health care for homeless people was started in Washington, D.C.—the Zacchaeus Clinic.

The clinic was funded entirely by donations from individuals, churches, community groups, and small grantors. It used a combination of paid staff and volunteers. It was established as a free-standing clinic in Washington's inner city as a response to the unmet needs of homeless people. People found out by word of mouth that they could receive health care with dignity and without waiting for long periods, as they often did in traditional outpatient departments and emergency rooms (Bargmann, 1985).

Although many health clinics have been developed in response to the needs of homeless people, they often also treat the domiciled poor, especially those who live in the immediate neighborhood. Other clinics were originally developed to serve poor people in general and now, for various reasons, find themselves serving increasing numbers of homeless people.

Specialized Health Care Approaches

Various other programs address the special needs of homeless people or the problems of specific subpopulations among the homeless.

Respite and Convalescent Care

One of the most serious issues facing those who work with homeless people is that many standard forms of treatment assume that the patient has a home; when that is not true, treatment is extraordinarily difficult. Convalescent (or respite) services allow a homeless person to recover from an illness or an injury that does not require (or no longer requires) care in a hospital but that is of such severity that the homeless person should not return to a regular shelter setting.

One example of a private effort is Christ House in Washington, D.C. As a result of a bequest, the Church of the Savior acquired and renovated an abandoned apartment house and converted it into a 34-bed respite facility; it has a paid and volunteer staff, including medical and nursing supervision and care. The Columbia Road Physician Group provides medical support, and all four doctors involved in the project live, with their families, on the top floor of the building, so that medical attention is available around the clock. When more intensive care is needed, local hospitals are used.

A similar program is the 40-bed respite unit at the Charles H. Gay Shelter Care Center for Men in New York City, which is a public-private effort. The shelter (including the respite unit) is funded by the New York City Human Resources Administration and is administered by the Volunteers of America, which is under contract with the city. The respite unit is adjacent to the on-site medical and nursing clinic administered by St. Vincent's Hospital (see above) and receives nursing support from the clinic. Referral for backup hospital services is either to St. Vincent's Hospital or to one of the hospitals of the New York City Health and Hospitals Corporation.

Residential Placement

Many homeless people with physical disabilities, mental disabilities, or both who cannot live independently require supportive living settings. One program that attempts to meet this need is the Veterans Administration (VA) community placement program, which secures supervised housing for mentally or physically disabled veterans who are facing discharge from a VA medical center and who would be at extreme risk of becoming homeless. Members of the committee visited four such placement sites in Lexington, Kentucky. Three were private homes in which the individual homeowner contracted with the VA to accept patients from the medical center (the largest program accepted up to eight men) for supervised residential living. The fourth program was a personal care home licensed by the Commonwealth of Kentucky. The personal care home received clients from the state agencies serving the mentally ill and the mentally retarded, as well as from the VA medical center. This facility is larger (over 15 beds) and was specifically designed to serve a population in greater need of medical and nursing care. Although the residences were supervised and certified by government agencies, the actual funding for the individual veterans comes from their own VA benefits.

In each of the programs identified above, communication and coordination were accomplished by individualized approaches developed over a period of time with systems that were more or less unique to each city. The programs were primarily targeted to the homeless; funding and other resources ranged from the purely charitable to the wholly publicly funded. However, a comprehensive, cohesive system of services is lacking. Even those programs that had strong ties with a hospital did not network with programs that serve, for example, the mentally ill or substance abusers.

The Johnson-Pew Health Care for the Homeless Projects

The most significant event to occur in the area of health care for homeless people in recent years was the creation of the Health Care for the Homeless grants, funded jointly by the Robert Wood Johnson Foundation of Princeton, New Jersey, and the Pew Memorial Trust of Philadelphia. In many respects, the creation of this joint program reflected the growth of the homeless problem and the fact that agencies that had historically been able to provide services to the homeless could no longer cope with their increasing numbers. The national Health Care for the Homeless program was developed to provide cities (applications were limited to the 50 largest cities in the United States) with an opportunity to make a significant impact on health care delivery to the homeless. On December 12, 1983, in a joint news release, the two foundations announced what was, in effect, the first attempt to address this problem on other than a local level. Cosponsored by the U.S. Conference of Mayors, the program guidelines required that cities forge a coalition of disparate groups of health care professionals and institutions, volunteer organizations, religious groups, public agencies, shelter providers, and members of the philanthropic community. This coalition was charged with developing a program to meet the health care needs of the homeless, improving their access to other supportive services and entitlements, and developing a strategy for continuing the program services after the termination of foundation funding:

As such coalitions strengthened and institutionalized their functions, it was hoped that they would become permanent structures for addressing the health and related needs of the homeless beyond the four year grant period. (Clark et al., 1985)

Of the 50 cities eligible for the program, 45 submitted grant applications; of these, 18 were funded under the national program and 1 city was funded under a special arrangement. A total of $25 million was allocated by the foundations, and each city received up to $1.4 million for use during a 4-year period.

One issue frequently raised as a result of the Johnson-Pew projects is whether it is necessary to develop separate health care systems for the domiciled and for the homeless. The answer to that question depends on the resources of an individual community and the willingness of existing health care systems to respond to the needs of homeless people. Even the most rigid system can, over a period of time, change to accommodate new programs; therefore, when it is necessary to develop parallel programs, it is frequently with the expectation that at some future time the newer program will be incorporated into existing programs. An especially good example of this is the incorporation of the Nashville Johnson-Pew project into the municipal health department.

Structure of the Johnson-Pew Projects

The 19 Johnson-Pew projects are distinctly different and highly idiosyncratic, and as such, they reflect the specific needs of the 19 cities in which they are located. The original request for proposal issued by the Johnson and Pew organizations required that all proposals be developed by a broad-based community coalition. Therefore, not one of the 19 programs was incorporated as a separate entity (although several are now in the process of seeking such incorporation). Each grantee needed to establish a system of governance and fiscal accountability, in effect, a fiduciary agent. The more common models provide for the funds to be authorized to an existing health care-related or social service-related agency (e.g., in New York City, the United Hospital Fund of New York; in Philadelphia, the Philadelphia Health Management Corporation) or to a charitable foundation (e.g., in San Francisco, the Episcopal Archdiocese of California; in San Antonio, the United Way). In two cities (Newark and Phoenix), the funds go directly to agencies of the municipal government. It is not yet possible to determine which funding methods are most effective.

In some of the projects, services are provided by staff who are employees of the project, with a single set of policies and procedures. Some programs rely on contracts with existing providers of health care services, to provide either specialized services (e.g, dental care) or general health care services to a specific geographic area (e.g., the New York City project has contracts with three existing health care agencies that provide services in different boroughs of the city). Staff are employees of the contract agency and are subject to the policies and salary schedules of that agency. Sometimes a mixture of direct and contracted services is provided. Certain services—such as case management—are provided directly by salaried staff, while other services—such as clinic operations—are provided by a contractor.

How services actually get to homeless people is probably the most varied (and creative) aspect of the Johnson-Pew projects. The methods of service delivery include mobile vans outfitted as clinics, mobile teams going to existing programs that serve homeless people (particularly shelters and soup kitchens), and central clinics located in areas where homeless people can be found in substantial numbers.

Common Elements of Health Care Programs for the Homeless

Although the Johnson-Pew projects are just past the midpoint in their 4-year grants, much has already been learned from these projects. As with the earlier models, there is no statistical basis to determine a program's success.5 In the course of its review and after many discussions with service providers as well as the homeless people who receive care, the committee identified the common elements that follow as especially significant.

Holistic Approach

Rather than treating an isolated health problem without considering the person's social or environmental situation, these programs provide treatments that recognize the interaction between the illness and the state of being homeless. The nature and level of the individual's entitlement benefits, for example, whether they are sleeping in the streets or in a shelter, where they get food, and so on, are taken into consideration in developing a treatment approach.


Health care is brought to areas where homeless people can be found. These targeted services can then serve as a conduit by which other services (including application and advocacy for entitlement benefits) are offered.

Empathetic Staff

Staff are aware of the attitudes that increase their effectiveness in working with the homeless population. In particular, staff recognize the exigencies of survival that impinge on the day-to-day activities of the homeless and the effects of those demands on the individual's health and health care.

Multidisciplinary Approach

Teams working with the homeless encompass a range of disciplines, including physicians, nurses, physician's assistants, nurse practitioners, and social workers. Given the range and severity of illnesses present among the homeless population, when volunteers are used (especially medical or nursing students), proper supervision is provided.

Case Management and Coordination of Services

One of the most critical elements in serving the homeless involves the coordination of patient treatment and the provision of access to other health care and social services with the aim of breaking the cycle of homelessness. The most frequent approach is to include social workers as part of the multidisciplinary team. This individual keeps in touch with service providers at other treatment sites to ensure that the homeless person follows through with the treatment plan.


Homeless individuals have few other people to rely on, often leading them to be very distrustful of people in general. The continuity of the program staff helps to build trust. Changes in personnel or disruptions in schedules increase their wariness; conversely, seeing familiar faces (especially if they can be treated by the same people both at an outreach site and then again in a clinic setting) increases their cooperation.

Range of Health Care Services

The final element is that successful programs offer a broad range of services. For example, they make some provision for convalescent care, prenatal care, and treatment for alcohol problems. Access to services such as convalescent care on discharge helps to prevent unnecessary rehospitalization. Not all programs are able to provide this range of services directly; however, they do recognize the service gaps and try to fill them in by working with other service providers.

The Health Care for the Homeless projects in the cities of Milwaukee, Nashville, and Detroit illustrate some of the structured and programmatic elements developed by the several Johnson-Pew projects. In Milwaukee, the project is administered by the Coalition for Community Health Care, Inc., a nonprofit organization established in 1979 to advocate the health care of indigent people. The Milwaukee project chose to contract with social service agencies for the social service component of the project and with four medical facilities for health care services. The medical facilities provide pharmaceutical and medical supplies, as well as x rays and laboratory services. The county hospital provides optometry services and the local community health centers provide podiatry services. Dental care is provided by the community health centers. By using multiple health care facilities, the cost of those services and supplies that were not funded by the Johnson-Pew projects is more evenly distributed. The Milwaukee project uses mobile teams that go to sites at which homeless people receive other (nonmedical) services. Paraprofessional outreach workers approach homeless or alcoholic people on the streets and assist them in obtaining financial entitlements and then permanent housing. The progress of these people is then followed as they participate in support groups for substance abuse and maintenance on medication.

In Nashville, one of the smaller cities in the project and one with even fewer services for homeless people, the original grantee organization was the Council of Community Services, Inc., a social services coordinating agency. The project established a stationary clinic, the Downtown Clinic, which was initially located between the two major shelters for homeless people. Staffed mostly with midlevel practitioners, including a physician's assistant and a family nurse practitioner, the clinic also has a full-time medical director who is a National Health Service Corps doctor. Clients come to the clinic because of its proximity to where they are sheltered and because of word-of-mouth recommendations. The clinic itself is a traditional outpatient facility, including an x-ray suite, which enables the staff to provide diagnostic work immediately rather than referring the clients elsewhere. Besides the stationary clinic, there is outreach to homeless people on the streets, on the river bank, in the shelters, and at various service agencies. Since the project began, the Meharry Community Mental Health Center and the Tennessee Department of Mental Health and Mental Retardation have provided funding for two mental health care professionals to work with the Johnson-Pew project, primarily in outreach to homeless chronically mentally ill people on the streets. As indicated earlier, the Nashville project will continue after Johnson-Pew funding has run out through incorporation into the Metropolitan Nashville/Davidson County Health Department, which has now succeeded the Council of Community Services as the grantee organization.

In Detroit, the Health Care for the Homeless project is administered by the United Community Services of Metropolitan Detroit. The project contracts for some of its services and provides other services directly. Three clinical and two administrative staff are on the project's payroll, while other staff are paid for through contracts: a part-time clinical nurse on contract with the Detroit Department of Public Health and one part-time physician and an administrative staff member on contract with the Wayne State University School of Medicine. In addition, the Detroit Department of Public Health has provided the project with two full-time staff members paid for by that department: a physician and the project director. The project has both mobile teams and a stationary clinic. The mobile teams visit eight different sites (shelters and soup kitchens) during the course of a week. If a client needs a more thorough workup, a referral is made to the clinic, which operates one morning per week. The clinic is located in space at the Detroit Receiving Hospital that was made available at no cost; the hospital also provides the clinic with much of its supplies. While the project purchases most of its own medical supplies and pharmaceuticals, x-ray services are provided free of charge by the Detroit Department of Public Health and Detroit Receiving Hospital. Optometric care is provided through a unique partnership: Clients are referred by the project to the Detroit Optometric Institute for eye examinations and, if necessary, for glasses. The Optometric Institute bills the Fort Street Presbyterian Church, a downtown church with a history of helping the homeless, at a reduced rate and the church pays for all the optometric services. Podiatric services are provided by the Detroit Department of Public Health; emergency dental services are provided at no cost by the health department, while general dental care is provided by the dental school of the University of Detroit on a sliding scale basis. The project has received a grant from the state of Michigan to fund mental health care services. However, if the client is already known to one of the neighborhood mental health care centers, a referral is made to that center rather than attempting to provide direct services. In times of emergency or crisis, clients are referred to Detroit Receiving Hospital for inpatient admission.

The three programs described above use various funding and administrative mechanisms for the provision of specific services (e.g., a hospital for x-ray services, a dental school for dental services, and a mental health center for psychiatric services). Multiple administrative arrangements place a more serious burden on the role of case management than if those services were all under one roof and provided by one agency. For example, in Detroit the Johnson-Pew project provides case management to its clients. However, the success of case management depends on feedback from the staff of the facilities to which the Johnson-Pew team refers clients; such feedback is not always forthcoming. It should also be mentioned that the clients of these projects are, primarily, individual adults. The provision of such services to families presents a different set of problems.

The three project examples illustrate the point that success does not necessarily depend on the form of governance (whether it be by a not-for-profit board, as part of an existing health care or social service agency, or by an established charitable foundation), the manner of administration (whether by direct provision of services or by contracting with existing service providers), or the method of operation (a centralized clinic in an area where homeless people congregate or mobile teams going from one area of a city to another). In fact, these modes may or may not apply to other cities; structure and administration seem to be most effective when they reflect the individual characteristics of a specific city. What appears to be most significant are the presence of all—or at least most—of the seven common elements of health care programs for the homeless described above.

Issues Raised by the Johnson-Pew Project Models

Stationary Clinics or Mobile Teams

Each of these models has a somewhat different approach to providing health care services to homeless people. The common denominator, however, is a permanent outreach component. In some instances, a stable clinic site may be developed that has outreach workers, and in other cases the mobile teams go to places where the clients congregate for services.

Each of the different components of these models has strengths and weaknesses. For example, stationary clinics are able to provide more sophisticated technology (e.g., x-rays and electrocardiograms [ECGs]). On the other hand, such clinics might be viewed as too threatening for clients, especially if these individuals have had negative experiences with emergency rooms or clinics in the past. The informality of shelters and soup kitchens may be perceived as less threatening than an outpatient clinic; therefore, mobile teams may be in a better position to overcome such initial reluctance. Guests at these sites often ask for blood pressure checks or adhesive bandages as a means of testing the health care professionals. However, such sites do not usually provide ancillary services, such as ECGs. Many of these sites reach clients who would not otherwise have access to care; unfortunately, the sites are somewhat less stable: Soup kitchens may close down for a week or a month because of a lack of funding or to give volunteers a rest; there is sometimes difficulty finding space to set up a temporary clinic with access to running water; soup kitchens and meal programs are often open only for a brief period of time in the middle of the day or in the evening; and programs for the homeless sometimes close in April and do not reopen until November. Although there may be large numbers of guests, the patient flow and the abbreviated hours make it difficult for the efficient and effective use of staff.

Small Social Service Agencies or Large Health Care Facilities

Locating health care services in a social service agency may often mean that ancillary services or specialty clinic appointments must be negotiated on a case-by-case basis, which is a time-consuming procedure. Moreover physicians, nurse practitioners, and physician's assistants might find it professionally isolating to practice health care outside of a health care facility. Recently, a New York City mobile health team for the homeless was transferred administratively from a very small neighborhood health care center to a large teaching hospital. While the move did not affect to any substantial extent the provision of direct services to their clients (they continued to receive the same services at the same sites), it was welcomed by team members because of the additional supports that a large facility could provide.

Direct Provision of Services or Contracting Services

The model of contracting of services, while appropriate in many instances, may pose specific problems. If the contracted staff are employed by a major health care agency, they may experience a division of loyalties between their employer and the program that specifically serves the homeless. Such contract agencies often have different personnel policies in terms of holidays, vacations and pay scales, opportunities for advancement, staff development programs, and so on. These differences can negatively affect the morale of team members who work for different employers.

Size of Areas To Be Served

Another area of concern in developing models for delivering health care is the size of the geographic area to be served. In some JohnsonPew cities, services have been focused in one downtown area where there may be the heaviest concentration of homeless people, effectively excluding large numbers of homeless people outside that catchment area. Other cities have tried to cover as broad an area as possible, which may result in the loss of valuable service delivery time because of the amount of time spent in transit. If the area covered is substantial, a sizable number of backup agreements may be necessary to ensure the provision of ancillary and diagnostic laboratory services for each treatment site. To a great extent, the combination or range of services and the geographic size of the area to be served are dependent on the amount of resources available.

Targeted Services for Populations with Special Needs

Concurrent with the development of the Johnson-Pew Health Care for the Homeless projects, other new forms of services to the homeless have been introduced. These programs provide health care to specific subpopulations among the homeless such as the chronically mentally ill, adolescents and youth, and homeless people with AIDS (acquired immune deficiency syndrome). Not only do they represent a growing recognition of the heterogeneity of the homeless population, they also indicate an increasing awareness that specific populations among the homeless require specific health care services designed to meet their particular needs.

The Chronically Mentally Ill6

Studies of the mental health of homeless people indicate that the prevalence of serious mental disorders is considerably higher among the homeless than it is among the general population (see Chapter 3). The provision of mental health services to homeless people is made difficult primarily by the lack of appropriate facilities and resources and by their extreme poverty, their lack of insight into their psychiatric problems, their distaste for psychiatric treatment, and the complexities of their service needs. Those needs, therefore, are often poorly met (Lamb, 1984).

As with homeless people in general, the seven elements delineated from the experience of the Johnson-Pew projects (a holistic approach, access and outreach, empathic staff, interdisciplinary approach, case management, continuity of services, and broad range of services) are also important in creating services for the homeless chronically mentally ill. To ensure effective treatment for this group, specialized forms of housing (including a range of supportive services, from independent living with minimal supervision to round-the-clock supervision in a community residence) and rehabilitation programs are also essential.


For severely mentally ill people, specialized housing arrangements are needed. In the wake of deinstitutionalization, various supervised housing arrangements have been developed. Model programs such as Fountain House in New York, a private, not-for-profit agency, combine a treatment center with supportive living in nearby apartments. With respect to public programs, one recent effort is the attempt of the New York State Office of Mental Health to provide supportive housing ranging in size from small community residences of fewer than 15 beds to the proposed larger residential care centers that resemble SRO hotels. Such centers, although funded by the state, are generally operated by well-established not-forprofit agencies. One of the requirements for funding is that the sponsoring agency must guarantee the provision of full-day treatment services for all residents of the facility, either directly or by means of contracts with other agencies. In addition, for those individuals whose disabilities are so severe as to warrant it, such services must be provided on-site until such time as the individual is able to negotiate community systems.


Structured psychosocial rehabilitation programs are often necessary to enable mentally ill people to function at their maximum capacity in the community. For the homeless, in most cases, the very structure of such programs may be a deterrent. Therefore, the principles of psychosocial rehabilitation must be brought to the shelters or residences. Over time, this should enable the individual to move from the on-site setting to a community setting. Assistance is also needed for homeless people to regain lost skills in the activities of daily living (ADL) that most of us take for granted (e.g., personal hygiene, cleaning, and basic meal preparation). The Community Support Systems program at the Volunteers of America shelter in New York is an example of such an effort. It provides a model apartment as part of an ADL skills training program. A similar program exists in Phoenix, where six apartments are used to help chronically mentally ill homeless people learn the skills required for independent living before moving into such situations.

A comprehensive array of services is needed by chronically mentally ill homeless people, but in most communities a full range of services does not exist. Until comprehensive service systems for mentally ill homeless people are developed, the care that clinicians can give to these patients is limited.

The committee and staff visited several programs for the homeless mentally ill. In addition to serving as potential models for services to chronically mentally ill homeless people, they are also possible solutions to the problems of other subpopulations. Some of these programs are described below.

Robert Wood Johnson Foundation Project for the Chronically Mentally Ill

The original Johnson-Pew Health Care for the Homeless program did not include specific services for homeless people who are mentally ill. Subsequently, the Robert Wood Johnson Foundation initiated a program to address the issue of providing services to the chronically mentally ill (including the homeless) in early 1986. It is jointly sponsored by the U.S. Conference of Mayors and by the U.S. Department of Housing and Urban Development, which has committed $75 million in additional support for Section 8 housing certificates. This program, administered by the Department of Psychiatry of Harvard Medical School, seeks to develop systems of coordinated services to the chronically mentally ill in the nine cities that receive funding under its auspices. The program attempts to bring disparate funding and organizational systems into a unified structure, so that the chronically mentally ill person need not go from place to place to receive services and run the risk of neglect. In addition, the housing component is designed to support the clinical component by providing housing appropriate to the individual's level of disability.

Self-Help Centers

During their site visit to the San Francisco area, several committee members toured the Oakland Independence Support Center. This is a day treatment center for homeless mentally ill people that is operated and staffed exclusively by people who themselves have been treated (either as inpatients or as outpatients) for mental illness. The center is run on the basis of group decisions regarding programs and has proved to be especially effective in obtaining housing for its ''members," primarily because it is willing to assist both the tenant and the landlord in resolving problems and disputes. Because of this willingness to maintain extensive involvement with members and to intervene to resolve such problems, landlords tend to seek out people involved with the center as tenants. Similar self-help centers have been developed in other cities. Self-help centers such as this one are partly dependent on the leadership of an individual or a small group.

Outreach Street Teams

A third approach for providing services to the homeless mentally ill is the development of outreach teams specifically seeking to identify homeless people in need of services. As indicated before, the Nashville Johnson-Pew Health Care for the Homeless (HCH) project has been supplemented by funding from the Tennessee Department of Mental Health and Mental Retardation's community initiative program that provides two case managers to walk the streets, developing rapport with chronically mentally ill homeless people. A similar effort is directly funded by the Johnson-Pew project in Los Angeles; there, the outreach worker was himself once a homeless person. The Volunteers of America (VOA) outreach program at transportation facilities of the Port Authority of New York and New Jersey also makes extensive use of former homeless people as members of their teams. The VOA program is jointly funded by an interstate compact agency and by a department of the local government (the New York City Department of Mental Health). New York is also the site of one of the most highly publicized programs, Project Reachout, which sends staff into Central Park to make contact with homeless people who literally live in the park (Goddard-Riverside Community Centers, 1986).

Veterans Administration Homeless Chronically Mentally Ill Program

One of the most recent and most extensive efforts to address the needs of the mentally ill homeless is the Homeless Chronically Mentally Ill (HCMI) program operated by the VA as mandated by P.L. 100-6. The law was enacted on February 12, 1987, and the program was operational within 4 months, an accomplishment made even more notable by the fact that this program encompasses 43 sites in 26 states. In its first 4 months of operation (May to September 1987), the outreach staff made contact with 6,342 homeless veterans. In designing the program, the staff of the VA's Division of Mental Health and Behavioral Sciences made extensive use of the experience of some of the programs previously described in this chapter, especially the Johnson-Pew HCH projects 7 (Rosenheck et al., 1987).

Briefly, the program is operated out of 43 (of the 172) VA medical centers throughout the country. Outreach teams of two or more staff work in the streets and at community sites (e.g., shelters and soup kitchens) at which homeless people congregate; the HCMI staff identified such unusual points of contact as the middle of a sandbar in a river in Oregon and under a bridge in Indianapolis. Of the total number of contacts made, 64 percent took place in the community, 30.2 percent at VA medical centers, and 3.5 percent at veteran centers; 61 percent of the contacts were initiated by HCMI staff, 17 percent by the veterans, and 12 percent by referral from community-based programs.

The basic structure of the program is as follows: outreach contact, intake assessment, psychiatric and medical assessments, linkage with existing providers of services (both VA and non-VA), residential treatment if necessary, and comprehensive community-based mental health treatment if necessary. As is often the case with such programs, there is a steady drop in numbers as homeless people move through such a process. Of the 6,342 homeless people with whom contact was made during the first 4 months of the program, only 4,010 completed the intake assessment phase; of those, only one-third completed the psychiatric assessment and only one-fifth completed the medical assessment. This can be attributed in part to the fact that the psychiatric and medical assessments were performed primarily at VA medical centers (sites to which many homeless veterans are wary of going); the VA reports that it is actively pursuing alternate sites in the community. Despite these problems, the VA reported 360 placements in residential treatment during the same period.

The VA HCMI program has several features worth noting. First, early in the program the decision was made to include those "at serious risk" of homelessness among those who would be eligible for services. Of those veterans for whom an intake assessment was completed, 8.7 percent were in intermittent residence with family or friends and 8.5 percent had a room or apartment of their own from which they were in danger of eviction. This makes the HCMI programs one of the few such programs to address formally the issue of preventing homelessness.

Second, due to the relatively isolated location of many of the 43 VA medical centers included in the program, the VA made a policy decision to contract with non-VA programs to provide residential treatment at sites near where the homeless veterans are found. In addition, provision has been made to offer community-oriented case management during and after such treatment to stabilize such placements.

Finally, prior to start-up and continuing through implementation, the VA provided training to the outreach workers, first at four regional training sessions conducted in Alabama, California, Missouri, and Virginia. This training involved such sophisticated techniques as videotaped demonstrations and role playing. Subsequent follow-up included site visits by staff from the VA central office in Washington, D.C., to staff in the field and monthly telephone conference calls with staff at all 43 sites.

Alcohol and Drug Abuse

As with the mentally ill, a homeless person who is an alcohol abuser is unlikely to benefit from an approach that does not include a range of services. Nonalcohol substance abuse is a more recent phenomenon and appears to be limited somewhat to younger homeless people (under age 40) and to certain cities, primarily on the East Coast. Again, this population needs a full range of services, including specialized housing. The committee was able to identify several programs targeted toward homeless people with alcohol or drug problems. The Guest House in Milwaukee serves a homeless population that is evenly divided between the chronically mentally ill and chronic substance abusers. Harbor House in St. Louis is an alcohol rehabilitation program for homeless men; health care services are provided to Harbor House by a local voluntary hospital.

The following are some examples of programs that have been described in the professional literature.

  • Case management teams, such as those funded by the Illinois Department of Alcoholism and Substance Abuse.
  • Shelters specifically for homeless people who are alcoholics or drug abusers; the Illinois Department of Alcoholism and Substance Abuse funds three such programs, as does the Massachusetts Division of Alcoholism and Drug Rehabilitation. The Maine Office of Alcoholism and Drug Abuse Prevention funds four shelters, one of which is also the site of a bakery that is used to provide training and employment for the homeless clients.
  • Programs for permanent or long-term housing, such as those funded by the Massachusetts Division of Alcoholism and Drug Rehabilitation; alcohol-free housing programs developed in Los Angeles, Seattle, and Portland, Oregon. In addition,
  • Intensive residential programs that combine treatment on-site; one example of this is a program in New Jersey, funded by the state Division of Alcoholism. Under this program, a small (10-person) group home for homeless alcoholics is also the setting for day activities (e.g., counseling, vocational training, education, and leisure time activities), as well as referrals for health and mental health care services.

These examples indicate that programs can be developed to address the needs of homeless people who have problems with alcohol or drugs.


It has been only 8 years since AIDS was first publicly identified. Since then the disease has reached epidemic proportions, and services for people that have been affected have not kept pace with demand. This is especially true for homeless people with AIDS. During a site visit to San Francisco, members of the committee met with the AIDS Advisory Committee of the San Francisco City/County Department of Public Health, as well as the AIDS discrimination specialist of the San Francisco Human Rights Commission. That meeting basically confirmed that those issues described in the Institute of Public Services Performance, Inc., report in New York (see Chapter 2) are not limited to any one community. The advisory committee reported that such matters as loss of housing due to loss of employment, discrimination, or while awaiting entitlement eligibility determinations are very real problems.

Members of the committee toured the Folsom Street Hotel program in San Francisco, a residential program for homeless people with AIDS operated by Catholic Charities, with partial funding by the City and County of San Francisco. Although it did not tour the facility, the committee is also aware of Bailey House, a similar program in New York City that is operated by the AIDS Resource Center with funding from the New York City Human Resources Administration and a grant from the U.S. Department of Health and Human Services. Both of these programs were developed to provide housing and supportive services to this population. However, because many patients with AIDS are increasingly disabled as the disease progresses, additional forms of housing with varying levels of physical care, up to and including skilled nursing facilities and hospices, are needed. The alternative seems to be hospitalization, which is extremely expensive.

Homeless Individual Women, Families, and Youths

Adult Individual Women

Other subpopulations among the homeless have been identified in recent years as needing specific targeted approaches. Members of the committee toured several programs for homeless individual adult women. Site visits were conducted to the House of Ruth, a shelter for homeless women in Washington, D.C., and the Firehouse Annex, a drop-in center and transitional residence for homeless women in Chicago. While neither of these programs is specifically a health care program, both are clinic sites for their respective city's Johnson-Pew HCH projects. Another issue in evaluating services for homeless women is the high reported prevalence of mental illness among them. The committee toured a program for homeless emotionally disturbed women at the YWCA in San Diego, a program funded in part by the San Diego County Division of Mental Health Services.


The committee visited programs for homeless families in several cities, in particular Pilgrim House in Kansas City and Project Hope in Boston. These two programs are not specifically health care programs; however, Pilgrim House receives on-site health care services from Kansas City's mobile homeless health care team, and Project Hope receives services from the Boston Johnson-Pew family team. Another program that the committee toured was the Emergency Lodge in St. Louis, a large shelter for homeless families operated by the Salvation Army. Among the services provided to the families are daily health screening sessions and health education programs conducted by a public health nurse and a weekly clinic conducted by a volunteer health screening team. This program is unique in that it also has a free day care center, which enables homeless mothers to search for jobs during the day.

One program that specifically addresses the health care needs of homeless families is the Venice Family Clinic in Los Angeles. This is a free clinic that has been serving the Venice Beach/Santa Monica area for more than 17 years; its program has been augmented by the Los Angeles JohnsonPew project to allow for services to homeless families. St. Anthony's Clinic in San Francisco, operated by the St. Anthony Foundation, serves both homeless families and homeless individual adult men; however, in response to fears expressed by homeless mothers about single men, the clinic has separate entrances and treatment suites for each group. Although these two programs certainly are excellent examples of what can be done, actual programs specifically serving homeless families do not exist anywhere in the numbers needed, especially in light of reports that this is the fastest growing subpopulation among the homeless.

Homeless Youths

The three studies of homeless youths cited in Chapter 1 (Boston, New York, and Toronto) each reported on shelter populations. Despite the differences among the study populations, the similarities in the findings regarding the physical and mental health needs of this population are even more significant. All three studies reported that supportive services, rather than the provision of housing alone, are needed. Each also reported that both the number of attempts to run away and the length of time that the teenager has been away or in shelter(s) are important. The greater the number of attempts to run away and the longer the adolescent is in an institutional setting, the more difficult it becomes to place that youth in a noninstitutional setting or, having made such a placement, for that youth to remain there. While the three studies disagreed on the form that noninstitutional services should take, they agreed that, given the problems of this population, specialized services for this group are necessary.

The study by the Greater Boston Adolescent Emergency Network (1985) of Massachusetts emphasized the changing role of shelters in the network of youth services.

In the 1960s and 70s a network of emergency shelters was developed across the country to house an increasing number of "true runaways," young teens who had fled from a horrific home environment and had no place to sleep. From the phone booth on the corner or from a poster on a bus station wall, they were able to locate the nearest shelter and find a safe haven. Today, this still happens, but it is the exception, not the rule. . . . Due to the lack of other resources to accommodate and treat the chronic system youth, emergency shelters have become 30 to 45 day warehouses for adolescents with no place to go. Twenty percent (20%) of our sample were referred to shelter care not to manage a crisis, but specifically to shelter the youth while long-term care was being arranged. An additional 15 percent were referred from another temporary shelter.

Members of the committee toured the Larkin Street Youth Center in San Francisco. This program, funded in part by federal, state, and local governments and in part by charitable donations, serves runaway and throwaway youths in the Tenderloin section of that city. It provides services, including health and mental health care services, in a drop-in setting. The center reports a 74 percent success rate in getting youths off the streets; of those youths that have been helped, 40 percent have been returned to their families and 60 percent have been placed in foster care, usually in the communities from which they originally came. The center also has a very strong AIDS education program, a critical issue because so many of the adolescents have become involved with intravenous drug abuse, prostitution, or both.

In addition to the common element of being aimed toward specific subpopulations, many of these programs also exist as a result of joint public and private support. Some, such as the Nashville outreach program and the San Diego YWCA program, receive support from their state mental health agencies. Others, like Bailey House in New York and the

Folsom Street Hotel program in San Francisco, receive funding from municipal agencies. Still others, such as the Medeiros Center in Boston, receive funding from specific federal programs. Finally, some programs, such as the Larkin Street Youth Center in San Francisco, receive funding from several levels of government or, as is the case with the Volunteers of America outreach program in New York City, from specialized public authorities. What is noteworthy is that in each case additional resources have been forthcoming from the charitable sector, representing a true public and private cooperation.

Other Issues in Health Services for the Homeless

Coordinated Efforts in Non-Johnson-Pew Cities

In addition to inquiring into cities that have received grants from the Johnson-Pew Health Care for the Homeless project, the committee sought information as to how health and mental health care is being provided to the homeless in cities that did not receive such grants. Site visits were made to three cities that applied for the grants but that were not funded (Kansas City, St. Louis, and San Diego) and one city that, because it was not among the 50 largest, was not eligible (Lexington, Kentucky). Returning to the four elements that appear to be common throughout most programs providing treatment to the homeless, the committee was able to make several observations.


Notwithstanding the lack of specific funding for health care programs for the homeless, each of the four cities evidenced effective communication networks. Even though Kansas City and St. Louis were not successful applicants for the grant funding, the coalitions that were developed during the preparation of their applications continue in existence. This enabled the process of communication developed for the grant proposal to proceed further; each city chose to attempt (with some success) to implement the original grant proposal with other sources of funding. In San Diego there have been several successive coalitions and task forces that have enabled communication networks to expand, often with the city and the county governments taking an active role. In Lexington, which was not eligible to apply for a grant and therefore did not specifically need a broad-based coalition for this purpose, such a coalition exists nonetheless because of close cooperation among provider agencies and a very supportive municipal government.


Partly because they decided to proceed along the lines of their original grant proposals, Kansas City and St. Louis have achieved a reasonable level of coordination of services. In Lexington, in part because of its small size (population of approximately one-half million), coordination is less formal, but no less effective. San Diego, on the other hand, faces a serious problem arising from a separation of areas of responsibility mandated by the state of California: the county has responsibility for health care, mental health care, and social services and the city has responsibility for housing and public safety. Although the Johnson-Pew grant has enabled two other California cities (Los Angeles and San Francisco) to overcome this problem to some extent (especially in San Francisco, which has a combined city and county government), San Diego still faces serious problems of coordination of services.

Targeted Approaches

Each of these cities attempted to develop targeted services, especially health care services. Lexington has placed services for the homeless in one area in the downtown section of the city. St. Louis has residential programs near the psychiatric hospital that serves many chronically mentally ill homeless people. Kansas City is currently attempting to target services to homeless families, a problem that appears to be exacerbated by the economic decline of the surrounding farm communities. The San Diego County Department of Health Services has recently provided some funding for a clinic operated in the downtown area in which homeless people tend to congregate. None of these cities, however, has been able to develop the kind of specialized services that are provided by many of the Johnson-Pew projects.


In three of these four cities, attempts to provide health care services to homeless people have been with joint public and private funding, with the bulk of the public funding coming from the city governments. In San Diego, however, the health care programs are mandated to the county government, with the result that program locations are distant from the areas where homeless people are found, transportation is difficult to obtain, and it is time-consuming to travel to the programs; these are serious barriers to access.

The experiences of these four cities support the fact that specific efforts are needed to deliver health care services to the homeless and to earmark funds for such efforts. Even where programs do exist for homeless people, the lack of access and the absence of a targeted effort may vitiate them.

Range of Health Care Services

The experience of the Johnson-Pew projects and other providers of health care services for homeless people suggests that a wide range of services is needed. The range and the extent to which each service should be developed in a given city may be based upon such factors as the numbers of homeless people and the proportions of the various homeless subpopulations. An assumption of these health care services is that provision of social services is an integral component of health care.

Although many of these services are appropriate for all people (homeless or not) and are especially important for the medically indigent, they are of even greater importance to homeless people because of the high level of debilitation seen in that population. The following range of services could be considered basic primary health care for homeless people.


Outreach to people where they are, including the streets.


General medical assessment and treatment for chronic and acute illnesses.


Specific screening, treatment, and follow-up for such health problems as high blood pressure.


Pediatric services (including well-baby clinics, immunizations, and screening for lead poisoning) and diagnostic and psychosocial intervention programs for both preschool and school-age children to address emotional disability and developmental delays.


Ancillary services (dentistry, podiatry, optometry, and specialized diets).


Access to mental health care and substance abuse services, including access to specialized housing.


Referral and access to convalescent care, as well as long-term medical and nursing care for catastrophic illness.


Gynecological services.


Prenatal care.


Educational services, primarily with regard to family planning and the prevention of sexually transmitted diseases (including the free distribution of condoms as part of AIDS education efforts).

Any health care providers also should take into consideration specialized mental health and substance abuse services. Unfortunately, health, mental health, and substance abuse have traditionally had separate funding streams, even though all three can interact with each other. This often blocks the delivery of services to people with multiple diagnoses. Models for the treatment of dual or multiple diagnoses for homeless people are rare (the Prevention Research Center in Berkeley, California, is currently researching the limited models of services for homeless substance abusers). The fact that they do not exist in great numbers does not, however, indicate that there is not a great need for such services; all reports received (both from the literature and from the site visits) suggest just the opposite.

Discharge Planning

Discharge planning is a difficult and complex task at best. General hospitals, mental hospitals, mental retardation facilities, correctional facilities, and the foster care system often are remote from the community into which the person is being discharged. However, all of these service providers are mandated to develop discharge plans for each client or patient, and even the best of plans can break down. Often, there is no follow-up to determine whether the plan works, and there can be an almost total lack of communication and coordination among institutions, communities, and the income support and service systems. The core of the problem, however, is that there are not enough options available on discharge. Acute-care hospitals are discharging people earlier, and homeless people have no adequate place to recuperate. There are only a few facilities that are minimally comparable to Christ House in Washington, D.C., and one cannot discharge a homeless person to home care if there is no home. It is true that there are insufficient appropriate options for discharging homeless people from acute-care hospitals; but networks of institutional providers, community-based service providers for the homeless, and the public social welfare offices could at least facilitate a more appropriate discharge than to the streets or to an inappropriate shelter. It is also highly desirable that shelter providers set aside beds that could be used for infirmary care or convalescence.

Similarly, mental health programs could plan for discharging homeless people to a supportive living residence with an appropriate level of care. The short supply of such programs makes it difficult to develop discharge plans, but more extensive planning before and follow-up after discharge might prevent a significant number of failed placements. Correctional facilities and parole officers could better coordinate and monitor more intensively an individual immediately following discharge, when that person is most likely to be unemployed and is at a higher risk of becoming homeless. As noted previously, however, the lack of a full range of community-based placements is the worst problem in discharge planning, so that clients, patients, or ex-prisoners are thwarted from achieving the most independent level of functioning of which they are capable.

One seemingly critical issue relates to the preparations for the discharge of an individual who may be eligible for Supplemental Security Income benefits. Although it is not adequately publicized, the Social Security Administration does have a program for early predischarge application and review, primarily directed to those in psychiatric inpatient facilities (U.S. Congress, House, Committee on Ways and Means, 1987). Every effort should be made to utilize this program so that disabled people who might be eligible for SSI could conceivably have the application approved concurrently with their discharge. Unfortunately, the Social Security Administration does not yet have a procedure for early application and review.

Case Management

Over the course of this study, the committee heard repeated references to case management. During its site visits, the committee met and spoke with a number of individuals identified as case managers. These people came from a wide variety of backgrounds, including social work, psychology, nursing, and in one instance, from the ranks of the homeless themselves.

How one views case management often seems to depend on the viewer's own past experience with the case management process, both personal and professional. Some see case management as the critical link that determines the success or failure of a program. Others see case managers as just another level of organizational bureaucracy that serves as still another barrier to the access of services.

The major problem appears to be that case management is ill defined and the role of the case manager is inadequately described. Fortunately, during the course of this study two publications were released that sought—from very different approaches—to resolve this problem. The Task Force on Welfare Prevention of the National Governors' Association (NGA) has released a report (1987) on welfare reform, Productive People, Productive Policies, that views the role of case management and case managers in terms of all human services. The COSMOS Corporation, under contract with the National Institute of Mental Health, has published a report, Intensive Case Management for Persons Who Are Homeless and Mentally Ill (Andranovich and Rosenblum, 1987). These two reports together provide a wealth of information for anyone who wishes to read a detailed analysis of this process and for those who work with homeless people.

The COSMOS report presents the following definition of case management:

Case management, as a mechanism for facilitating the access and movement of an individual through fragmented service systems, is viewed as an essential feature for effective service delivery to individuals who are homeless and seriously mentally ill. It attempts to ensure that a community support system is maximally responsive to the specific, multiple, and changing needs of individual clients.

Both the NGA and the COSMOS reports identify similar functions that constitute the structure of case management:

  • identification and outreach—determining who is in need of services and bringing them into the service delivery system;
  • assessment—determining the client's individual strengths and determining the needs that must be met;
  • service planning—developing a plan to meet those needs;
  • coordination and facilitation—working with the client and service providers to arrange for the actual delivery of services necessary to meet those needs;
  • monitoring—working with the client and service providers to determine whether each service provider (or all service providers, if there is more than one) is meeting its obligations;
  • evaluating—determining when and if changes in the service delivery plan are necessary and then negotiating and monitoring the implementation of those changes; and
  • advocacy—acting for or with the client in obtaining those services (including housing) that are needed, with one of the ultimate goals being that the client eventually becomes his or her own advocate.

The COSMOS report also identified ''intensive" case management as critical to working with chronically mentally ill homeless people. It defined intensive case management as a more aggressive approach for those most in need, especially in the areas of outreach and advocacy. In addition, the NGA report speaks to the need for certain qualities in those who are case managers:

  • communication skills, both with the client and with service providers;
  • knowledge of such things as rules, regulations, programs, and resources;
  • empathy with the client and the ability to assist in seeing the client's strengths and to capitalize on those strengths;
  • ability to identify the critical issues facing the client and to identify the appropriate resolutions for those issues; and
  • ability to hold others—both the client and the service providers—accountable for their performance.

Liability Insurance Coverage for Providers

The committee received several reports of programs that have encountered difficulties in obtaining malpractice insurance. Part of the problem may be that the programs are based in social service agencies rather than in health care facilities, but the committee is unaware of instances in which legal judgments have been rendered in favor of a homeless plaintiff in a malpractice action. Therefore, the basis for setting extraordinarily high premiums for or denying malpractice insurance is unclear. The committee also received reports that several not-for-profit organizations lost their liability insurance coverage or were charged very high premiums, causing them to either suspend or curtail operations for extended periods of time. At present, these reports appear to be scattered and do not seem to represent any specific pattern of denial of coverage. The potential negative impact of the loss of insurance coverage on the ability of the private sector to continue to provide services to homeless people cannot be ignored.

The potential lack of insurance coverage is of special concern with respect to the growing involvement of universities (especially medical, dental, and nursing schools) in the provision of health care services to the homeless. The committee observed the involvement of the schools of nursing of the University of Kentucky in the programs for the homeless in Lexington and of the University of California, Los Angeles, in the Johnson-Pew project in Los Angeles. In addition, the committee is aware of similar programs with the Johns Hopkins Medical Institutions in Baltimore and the Georgetown University Dental School in Washington, D.C. Problems with insurance—both malpractice and general liability—could effectively forestall such efforts.


Staff recruited to work with homeless people often have some special characteristics as well as professional expertise: They are willing to work against all kinds of odds and to provide services where the people are, leaving behind the more traditional and protected clinic and office settings. The ability to be innovative and flexible is important for working with the homeless. Staff may be open to the development of techniques that are different from those of the academic medical model and the usual adult outpatient clinics. The treatment of health problems is complicated by all the psychosocial problems experienced by homeless individuals and families. Some clients may be distrustful, rejecting, or hostile. The problems presented may often overwhelm the best trained, most experienced workers. Many of the adaptive, creative responses that homeless people develop for coping on the street may work against their being moved into a domiciled situation. Making such changes and adaptations may be overwhelming and frightening for homeless people to contemplate. Finding the innovative approach to engaging such clients and motivating them to try changes is the ultimate challenge of professionals who work with homeless people.

Of central importance to the task is each staff member's commitment and willingness to work as a part of an interdisciplinary team in which there is distinct professional expertise but also some fluidity in roles. In addition, although it is unlikely that anyone would be attracted to working with homeless people for material reasons, too often the salaries that are offered reflect society's tendency to stigmatize the workers in the same manner as it stigmatizes homeless people. Salaries can be made commensurate with work load and experience as well as competitive in the employment market. Specific and appropriate training of staff is desirable. Some staff may already have worked with homeless people but not in the context of health care services; others may be health care workers who have not worked with homeless people. Still others may have worked with a different population of homeless individuals. Training might include:

  • issues relating to the homeless, for example, the causes of homelessness, the subpopulations, and the health problems of homeless people;
  • orientation to the agency, including its policies, procedures, and opportunities for staff development;
  • supervision, including the medical and social service aspects of the program;
  • interview techniques or other means of assessing emotional problems;
  • crisis intervention techniques;
  • problems of working with the chronically mentally ill;
  • identification of and strategies for confronting manipulative behaviors; and
  • issues of case management, for example, other resources that are available and how a homeless person can access those resources.


Although homeless people are a diverse group, the nature of their life situations and the multiplicity of their needs lead to the conclusion that they would benefit from specific approaches in the provision of health and mental health care services. Programs have been targeted to the homeless in general; specific programs have been targeted to certain subpopulations that are delineated by the nature of their health problems, demographic characteristics that necessitate specialized approaches, or their location, such as in rural and suburban areas. Even when such specialized services are provided, the coordination of efforts with other existing services is essential. The goal should be to enable homeless people to have access to the range of services that already exist, thereby decreasing their need for specialized services. The ultimate goal is to resolve whatever problems prevent homeless people from becoming domiciled.


  • Andranovich, G. D., and S. Rosenblum. 1987. Intensive Case Management for Persons Who Are Homeless and Mentally Ill. Washington, D.C.: COSMOS Corporation.
  • Bargmann, E. 1985. Washington, D.C.: The Zacchaeus Clinic—A model of health care for homeless people. In Health Care of Homeless People, P. W. Brickner, editor; , L. K. Scharer, editor; , B. Conanan, editor; , A. Elvy, editor; , and M. Savarese, editor. , eds. New York: Springer-Verlag.
  • Breakey, W. R. In press. Mental Health Services for Homeless People. In Homelessness: A National Perspective. M. Robertson, editor; and M. Greenblatt, editor. , eds. New York: Plenum.
  • Brickner, P. W., editor; , L. K. Scharer, editor; , B. Conanan, editor; , A. Elvy, editor; , and M. Savarese, editor. , eds. 1985. Health Care of Homeless People. New York: Springer-Verlag.
  • Clark, M. E., R. M. Neal, S. L. Neibacher, and S. L. Wobido. 1985. A flexible approach to health services for the homeless: The National Health Care for the Homeless Program. Paper presented at the Annual Meeting of the American Public Health Association, Washington, D.C.
  • Fischer, P. J., and W. R. Breakey. 1986. Characteristics of the homeless with alcohol problems in Baltimore: Some preliminary results. Departmentof Health Policy and Management, School of Hygiene and Public Health, and Department of Psychiatry and Behavioral Sciences, School of Medicine, The Johns Hopkins University, Baltimore, Md.
  • Goddard-Riverside Community Centers. 1986. Project Reachout: Services and Advocacy for the Mentally Ill Homeless. New York: Goddard-Riverside Community Centers.
  • Greater Boston Adolescent Emergency Network. 1985. Ride a Painted Pony on a Spinning Wheel Ride. Boston: Massachusetts Committee for Children and Youth, Inc.
  • Lamb, H. R., editor. , ed. 1984. The Homeless Mentally Ill. Washington, D.C.: American Psychiatric Association. [PubMed: 6479924]
  • Robertson, M., and M. Greenblatt. In press. Homelessness: The National Perspective. New York: Plenum.
  • Rosenheck, R., P. Gallup, C. Leda, P. Leaf, R. Milstein, I. Voynick, P. Errera, L. Lehman, G. Koerber, and R. Murphy. 1987. Progress Report on the Veterans Administration Program for Homeless Chronically Mentally Ill Veterans. Washington, D.C.: Veterans Administration.
  • Task Force on Welfare Prevention, National Governors' Association. 1987. Productive People, Productive Policies. Washington, D.C.: National Governors' Association.
  • U.S. Congress, House, Committee on Ways and Means. 1987. Background material and data on programs within the jurisdiction of the Committee on Ways and Means. 100th Cong., 1st sess., March 6, 1987.



The term Johnson-Pew is not generally used to describe these projects. It is used in this report because the more commonly used name of the project, Health Care for the Homeless projects, could just as easily describe many programs that are not funded by this particular grant.


For a more detailed description of the St. Vincent's program, see Brickner et al. (1985).


The term combat zone came into common usage in the 1960s to describe a section of downtown Boston known as the site of strip shows, adult bookstores, and so forth. Because of its reputation as being the more "open" part of that city, it became attractive to street people. It is now undergoing commercial redevelopment.


Both the St. Francis House and SOME, as well as the Pine Street Inn discussed earlier, have expanded in the past 2 years as the result of additional staff provided by the Johnson-Pew projects in Boston and Washington.


However, the Social and Demographic Research Institute data derived from the client contact reports of the various projects do represent the first such diagnostic and utilization statistical data drawn from more than just a single local source; they also represent a potential base for future evaluations of program effectiveness.


This section is based in part on Breakey (in press) cited in Robertson and Greenbelt (in press).


For a full description of the HCMI program, see Rosenheck et al. (1987). For a brief description of the veterans seen in this program, see Chapter 1 of this report.

Copyright © 1988 by the National Academy of Sciences.
Bookshelf ID: NBK218235


Related information

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...