Health Problems That Cause Homelessness

Certain illnesses and health problems are frequent antecedents of homelessness. The most common of these are the major mental illnesses, especially chronic schizophrenia. As mentally ill people's disabilities worsen, their ability to cope with their surroundings—or the ability of those around them to cope with their behavior—becomes severely strained. In the absence of appropriate therapeutic interventions and supportive alternative housing arrangements, many wind up on the streets. Another contemporary example of illness leading to homelessness is AIDS. As the disease progresses and leads to repeated and more serious bouts with opportunistic infections, the individual becomes unable to work and may be unable to afford to continue paying rent. Other health problems contributing to homelessness include alcoholism and drug dependence, disabling conditions that cause a person to become unemployed, or any major illness that results in massive health care expenses.

One type of health problem in this category—about which the committee heard much during several site visits—is accidental injury, especially jobrelated accidents. Although such programs as Workers' Compensation were designed to prevent economic devastation as a result of workplace casualties, they often fall far short of what is optimal for many reasons, including lack of knowledge of the program by the employee, low levels of benefits under the program, and lack of benefits for "off the books" work and migrant farm labor. A case study illustrates the point:¹

Samuel Anderson arrived in New York City in 1985 from his native Oklahoma. He is 24 years old, educated through the 11th grade, and says he left his rural surroundings because there was no opportunity to work, ". . . there was no job with something ahead of it." He feels that his chances will be best in the "biggest town I know of." In New York, he is studying for a graduate equivalency diploma and supports himself as an evening security guard. His wages are enough to pay for a rented room in the borough of Queens. Five months after starting work, he scuffles with intruders and suffers gunshot wounds in his right leg and hand (he is right-handed). Mr. Anderson spends 2 weeks in the hospital after losing four pints of blood through his wounds. A vascular surgeon and a neurosurgeon repair his shattered hand during a 4-hour microsurgical procedure. In the meantime, his room in Queens (he is in a hospital in the borough of Manhattan, some distance away) is rented to someone else because of his absence and the concurrent lack of rent payment. After discharge from the hospital, he spends a few nights in a hotel. When his money runs out, he sleeps in a city park, finally coming to a shelter.

In addition to accidents, various common illnesses such as the degenerative diseases that accompany old age can also lead to homelessness:

James Barnam, now 62 years old, has worked regularly since age 17, but has never found a job with secure employee benefits. He has lived a marginal existence: adequate funds for food and a room in a single room occupancy hotel, but certainly not enough for savings. He is fired from his long-held kitchen job because he cannot see the food stains on the dishes; after working 2 days as a messenger, he is let go because items were delivered to incorrect addresses. Mr. Barnam has eye cataracts, a frequent accompaniment of older age and treatable with ambulatory surgery for those patients with health insurance. Mr. Barnam's marginal income entitles him to Medicaid benefits, but he is unable to negotiate the public welfare system and has no one to guide him through forms, appointments, and examinations. Upon losing his hotel room, Mr. Barnam goes to a shelter for homeless men after he is discovered at a bus station by outreach workers. However, even there, his health problem remains troublesome: he almost loses his bed because he fails to sign a daily bed roster he cannot see.

In each of these cases, employment was not secure, and the man lacked a network of family or friends. The fact that health problems precipitated homelessness underscores the relationships among health status, employment, social supports, and access to affordable housing.

Health Problems That Result from Being Homeless

Homelessness increases the risk of developing health problems such as diseases of the extremities and skin disorders; it increases the possibility of trauma, especially as a result of physical assault or rape (Kelly, 1985).² It can also turn a relatively minor health problem into a serious illness, as can be seen by the case of Doris Foy:

Doris Foy's varicose veins occasionally result in swollen ankles. When homeless, she sleeps upright, and her legs swell so severely that tissue breakdown develops into open lacerations. She covers these with cloth and stockings—enough to absorb the drainage but also to cause her to be repugnant to others because of the smell and unsightly brown stains. She is eventually brought to a clinic by an outreach worker. When the cloth and the stockings are removed from the legs, there are maggots in the wounds. She is taken to the emergency room of a hospital, where her wounds are cleaned.

Other health problems that may result from or that are commonly associated with homelessness include malnutrition, parasitic infestations, dental and periodontal disease, degenerative joint diseases, venereal diseases, hepatic cirrhosis secondary to alcoholism, and infectious hepatitis related to intravenous (IV) drug abuse.

Homelessness as a Complicating Factor in Health Care

For even the most routine medical treatment, the state of being homeless makes the provision of care extraordinarily difficult. Even the need for bed rest is complicated, if not impossible, when the patient does not have a bed or, as is the case in many shelters for the homeless, must leave the shelter in the early morning. Diabetes, for example, usually is not difficult to treat in a domiciled person. For most people, daily insulin injections and control of diet are adequate. In a homeless person, however, treatment is virtually impossible: Some types of insulin need to be refrigerated; syringes may be stolen (in cities where IV drug abuse is common, syringes have a high street value) or, sometimes, the homeless diabetic may be mistaken for an IV drug abuser; and diet cannot be controlled because soup kitchens serve whatever they can get, which rules out special therapeutic diets. The following case illustrates the various problems involved in treating a homeless man with another common chronic medical problem, hypertension:

Tyrone Harrison is black, 26 years old, and homeless because he cannot find a job. He wants to work in the shelter kitchen and waits 3 hours for a preemployment physical examination. He is friendly and describes himself as "very healthy." His blood pressure is 180/120. His smile disappears and he feels "cut down." Because he is homeless, he must deal with his illness, private and asymptomatic, in the public spaces of the shelter. He refuses to talk about high blood pressure with the fellows in the dormitory—it diminishes his macho image. He tells the nurse that his blood pressure reading must be a mistake. Three weeks later, after six contacts with the medical outreach worker, he confides that his cousin had been a dialysis patient because of hypertensive kidney disease. Weeks later, after several more visits to the medical team, Tyrone consents to medication for his persistently elevated blood pressure. His 2-week supply of pills are stolen 4 days later. An argument erupts in the dormitory and, in accord with routine regulations, Tyrone is put out of the shelter for 2 weeks. On his return to the shelter, his blood pressure is uncontrolled because he had no medication.

The cases described above exemplify not only how homelesshess complicates treatment but how burdens are placed on various parts of the social system and on the homeless persons themselves. Because he lacked any form of health insurance, Samuel Anderson did not receive rehabilitation therapy for his right hand, and as a result developed stiffness and had significant loss of fine and gross motor skills; he had to apply for permanent disability benefits. Doris Foy was admitted to the hospital, because the treatment for her leg ulcers, which consisted of elevating her leg and taking prescribed antibiotics, is impossible for a homeless patient. Not only does her hospital stay make a bed unavailable for someone else who might possibly be in more serious need of inpatient treatment, but it also means that the hospital will not be reimbursed for her treatment because under the present system of utilization review, cellulitis with leg ulcers is judged to be treatable on an outpatient basis, and therefore, inpatient treatment for this condition may not be covered by Medicaid.

Traumatic Disorders

Contusions, lacerations, sprains, bruises, and superficial burns are more commonly reported in the homeless population (TRAUMA in Tables 3-1 and 3-2). Results of a 1983 study indicated that approximately 30 percent of 524 homeless people treated in San Francisco over a 6-month period presented because of trauma-related injuries (Kelly, 1985). Homeless people are at high risk for traumatic injuries for a number of reasons. They are frequently victims of violent crimes such as rape, assault, and attempted robbery. In addition, primitive living conditions result in unusual risks; for example, the use of open fires for warmth predisposes them to potential burns.

Most of the findings in the literature, including those from the national HCH program, describe inner city homeless people. It is not known whether these observations can be extrapolated to the homeless people in rural areas. For example, during visits to rural areas of Alabama and Mississippi, committee members commented on the relative infrequency of traumatic disorders.

Disorders of Skin and Blood Vessels

Pustular skin lesions secondary to insect bites and other infestations are common among homeless people (SERSKIN in Tables 3-1 and 3-2). In addition, venous stasis of the lower extremities (i.e., poor circulation because of varicose veins) caused by prolonged periods of sitting or sleeping with the legs down predisposes homeless people to dependent edema (swelling of the feet and legs), cellulitis, and skin ulcerations. Although there is reason to speculate that venous valve incompetence would develop more frequently in homeless patients and lead to chronic phlebitis, data are meager. The term "peripheral vascular disease" (PVD in Tables 3-3 and 3-4) is frequently used to connote venous stasis; there is no clear evidence that arterial vascular disease is more prevalent in this population than in a nonhomeless population. Recurrent dermatitis (MINSKIN in Tables 3-1 and 3-2), which is possibly related to inadequate opportunities to bathe or shower and which is associated with infestations with lice and scabies, is prevalent among the homeless population. This form of dermatitis is frequently confused with bacterial cellulitis, since they both present with red, warm, tender skin lesions. This confusion may lead to inappropriate management. Moreover, homeless people do have an increased frequency of bacterial cellulitis and other pustular skin lesions. Finally, homeless people are at high risk of developing subcu-Collecting taneous abscesses, but this may be related in part to an increased prevalence of needle-stick infections from drug abuse.

Respiratory Illnesses

Acute nonspecific respiratory diseases (MINURI and SERRI in Tables 3-1 and 3-2) are commonly reported in populations of homeless people in shelters. Living in groups, crowding, environmental stresses, and poor nutrition may predispose homeless people to infections of the upper respiratory tract and lungs.

Tuberculosis has become a major health problem among homeless people (TB in Tables 3-5 and 3-6). Characteristically, this has been a disease associated with exposure, poor diet, alcoholism, and other illnesses that can lead to decreased resistance in the host. Substance abusers and the elderly are at high risk for developing tuberculosis. Immigrants from Third World countries also have an increased risk of infection (U. S. Department of Health and Human Services, 1980; Brickner et al., 1985). In a study of tuberculosis among homeless people in New York City in 1980 (Sherman, 1980), based on tuberculin skin test reactivity and subsequent case findings, 191 people were initially screened. Of these, 98 had positive skin tests and 13 had positive sputum cultures for Mycobacterium tuberculosis. Forty-four required either prophylaxis or treatment according to recommendations of the American Thoracic Society. Compared with nonhomeless populations, these homeless individuals had a very high frequency of skin test reactivity and positive cultures. Whether homelesshess alone led to the high prevalence of tuberculosis or whether multiple other predisposing factors were equally important is not obvious from the results of this single study. However, other studies performed in New York City and Boston between 1982 and 1986 confirm earlier observations and support the findings that homeless people have a greater prevalence of tuberculosis (Glickman, 1984; Centers for Disease Control, 1985; Barry et al., 1986; Brickner et al., 1986; Narde et al., 1986). Because tuberculosis is spread by personal contact, these infections pose a potential public health problem to occupants of shelters and to the general population.

Chronic Diseases

The proportion of adults seen more than once in the HCH clinics who suffer from various chronic illnesses (e.g., hypertension, diabetes, and chronic obstructive pulmonary disease) is high—41 percent—compared with 25 percent in domiciled outpatients described in the NAMCS data (Tables 3-3 and 3-4). The high prevalence of hypertension can be explained partially by age, race, and alcohol consumption; but homelessness makes the long-term dietary and pharmacological management of hypertension extremely difficult.

Similarly, compliance with recommended treatment regimens for cardiovascular and renal diseases, as well as metabolic disorders such as diabetes, is notoriously difficult for homeless people. For these reasons, many homeless people are referred to hospitals for inpatient care for the treatment of disorders that in nonhomeless people could be managed on an Outpatient basis.

Miscellaneous Health Problems

Foot problems occur with a greater frequency among homeless people. These include superficial fungal infections and calluses, corns, and bunions that are apparently the result of trauma from ill-fitting shoes. Homeless people suffer from many dental problems. Reports of poor oral hygiene, cavities, gingival disease, and extractions with no prosthetic replacements appear to be extremely common among homeless people (TEETH in Tables 3-3 and 3-4). These problems are also common among indigent patients in general who have limited or no access to dental care. Finally, various illnesses associated with increased mortality are related to environmental exposure, such as hypothermia and frostbite or hyperthermia (Olin, 1966; Brickner et al., 1972; Alstrom et al., 1975). These life-threatening problems are especially prevalent among alcoholic homeless people and those who abuse other drugs.

Mental Disorders

Many homeless adults suffer from chronic and severe mental illness. The visibility of mentally ill people has led to the creation of a stereotype for the entire homeless population; the earlier stereotype of the homeless alcoholic has been replaced in recent years with that of the mentally ill homeless person (Stark, 1985, 1987). One of the first reports describing the high prevalence of mental illness among the homeless appeared in 1978, when Reich and Segal wrote about the Bowery in New York as a "psychiatric dumping ground." They asserted that large numbers of psychiatric patients were being discharged from the mental hospitals and ending up on the streets, a theme that has since been echoed widely in the media and professional literature.

Not only can homelessness be a consequence of mental illness, but a homeless life may cause and perpetuate emotional problems. To sort out these variables, it is necessary to distinguish among the various categories of psychiatric disorders. The major mental illnesses, principally schizophrenia and the affective disorders (bipolar and major depressive disorders), are unlikely to result from the trauma of homelessness. Rather, they cause a level of disability and impaired social functioning in some people that, in the absence of adequate treatment and support, may lead to homelessness, which will then exacerbate these conditions (Fischer and Breakey, 1986).

Personality disorders are not considered "major" mental illnesses because reality awareness is maintained; nevertheless, these disorders are manifested by a person's long-standing inability to deal with the routine demands of living (e.g., as a parent, worker, or independent citizen). Deeply ingrained maladaptive behavior patterns, which usually begin during childhood or adolescence, interfere with a person's capacity to relate to others, limit a person's potential, and often provoke counterreactions from the environment. Personality disorders should not be seen primarily as a consequence of homelessness. Rather, because they impair a person's ability to cope with the demands of life and the expectations of society, they may contribute to the factors that cause certain people to become homeless.

Other psychiatric illnesses, such as the anxiety and phobic disorders and milder depressive reactions, can either be contributing factors in causing homelesshess or, more commonly, result from the stress of homelesshess. Becoming homeless is a psychologically traumatic event that commonly is accompanied by symptoms of anxiety and depression, sleeplessness, and loss of appetite. Sometimes, homeless people try to "medicate" these feelings away with alcohol or drugs.

Dementia is a progressive deterioration of mental faculties resulting from degenerative brain disorders, such as Alzheimer's disease; recently, it has been observed among some people with AIDS. It can also be caused by repeated small cerebral hemorrhages or traumas from diseases such as untreated hypertension or uncontrolled epilepsy; it is also a relatively common consequence of chronic alcoholism. Certain types of dementia, therefore, would be expected to occur more commonly in homeless people.

In interpreting research data on the psychiatric disorders suffered by homeless people, distinctions among the different diagnostic categories are important. Crude data, such as a history of psychiatric hospitalization (see Table 3-7), are of limited value in defining the prevalence of psychiatric disorders or in predicting needs for mental health services. Estimates of the prevalence of current major mental illness range from 25 to 50 percent of individual homeless adults (Bachrach, 1984). The most frequently reported figure—both in the literature and in the committee's site visits— was approximately one-third. In addition, during those site visits, members of the committee occasionally received reports of homeless people who manifested symptoms that might indicate the presence of mental retardation or other developmental disabilities. However, no studies of this problem could be located, so it is impossible to identify the extent to which mental retardation is present among homeless people.

TABLE 3-7

Treatment History of Homeless Mentally Ill.

Some researchers have used screening instruments, such as the Brief Symptom Inventory, the General Health Questionnaire, or the Center for Epidemiological Studies Depression Scale, to measure psychiatric disorders among the homeless. These include studies done in St. Louis (Morse, 1986) and Detroit (Mowbray et al., 1985) and one statewide sample of Ohio (Roth et al., 1985). Invariably, authors who use these instruments report rates of psychological distress in homeless people that are higher than those found in other population groups. These screening instruments provide nonspecific measures of psychological distress and not diagnosis. They are therefore of limited value in describing the nature of mental health problems and predicting the needs of homeless people for services.

Several studies have sought to clinically examine homeless individual adults in shelters (see Table 3-8). Two recent studies reported a high prevalence of substance abuse (alcohol and other drugs) and major psychiatric disorders among this population. Arce et al. (1983) examined homeless people in Philadelphia and diagnosed major mental illness in 40 percent of those studied. When substance abuse, personality disorders, and organic disorders were included among the diagnoses, the figure rose to 78 percent of those studied. Bassuk et al. (1984), in a similar study in Boston, reported major mental illnesses (mania, depression, schizophrenia) among 39 percent of those studied; when substance abuse and personality disorders were included among the diagnoses, the figure rose to 90 percent.

TABLE 3-8

Studies on Mental Illness Among the Homeless.

Reports of findings such as these have stimulated strong criticisms in some circles. Some critics object to emphasizing the psychopathology of homeless people as an explanation for their homelessness, noting that failure to provide community-based care and appropriate housing are the crucial factors; they are concerned about "blaming the victim" instead of focusing public attention on the societal problems and political choices that are responsible for the increase in homelessness. Methodological critics point to the lack of standardized clinical diagnostic methods and samples restricted to a single shelter, so that the subjects may not represent the overall homeless population.

One study that addresses both methodological objections is that of Farr et al. (1986), who interviewed a sample of homeless men from various settings in the skid row area of Los Angeles. Trained interviewers, using the diagnostic interview schedule developed for the National Institute of Mental Health (NIMH) (Robins et al., 1984), gathered data on 379 subjects. By this method they determined that at the time of the examination, 60 percent of the homeless men met the criteria for one or more current (within the past 6 months) mental disorders or substance abuse disorders. Of the total sample, 11.5 percent met the diagnostic criteria for schizophrenia, 20 percent met the criteria for major affective disorders, and 3 percent displayed severe cognitive impairment suggestive of dementia; 17 percent met the criteria for antisocial personality disorder, and 31 percent apparently abused alcohol or other drugs. Many respondents met the criteria for the diagnosis of more than one type of serious disorder. When added together, the total percentage of homeless people who apparently suffer from any mental illness or substance abuse problem was 83 percent, a figure similar to those found in the studies by Arce et al. (1983) and Bassuk et al. (1984). Several years earlier, using the same diagnostic method, a randomly selected sample of domiciled people in Los Angeles was surveyed as part of the NIMH Epidemiological Catchment Area (ECA) program (Regier et al., 1984). Farr and colleagues (1986) found that, compared with domiciled men in the catchment area study, homeless people were given a current diagnosis of schizophrenia 38 times more frequently, major affective disorders 4 times more frequently, antisocial personality disorders 13 times more frequently, dementia 3 times more frequently, and substance abuse disorders 3 times more frequently.

A study of men at a mission shelter in Baltimore also used the diagnostic interview schedule (Fischer et al., 1986). As in Los Angeles, data from an ECA program sample of a domiciled population were available for comparison. Disorders in all categories occurred with greater frequencies in the 51 homeless men. Overall, 24 of the men (47 percent) had some current (within 6 months) mental or substance abuse disorder; 4 percent of the total sample suffered from major mental illnesses, and 20 percent suffered from alcohol or drug abuse problems. Using virtually identical standardized diagnostic procedures—but very different sample populations—these two 1986 studies (Farr et al. and Fischer et al.) demonstrated a high overall prevalence of psychiatric disorders in homeless men. It is not clear why the prevalence of specific major mental illnesses (schizophrenia, major affective disorder) was so much lower in the sample of Baltimore mission users compared with that in the sample of homeless people studied in Los Angeles—and elsewhere (Table 3-8). Differences in sampling strategy are the most likely explanation.

Although there is a substantial proportion of women in the homeless population, there are no data on the psychiatric diagnoses of randomly sampled homeless adult women. There is an impression, however, supported by nonspecific indicators of mental health status, that larger proportions of individual homeless women than homeless men are mentally disturbed. Researchers in New York City's shelter system (Crystal and Goldstein, 1984) reported that indicators of psychiatric disorder, such as history of hospitalization or current symptoms, were reported by 37 percent of women but by only 21 percent of men. Similar findings have been reported for homeless people from Milwaukee, Phoenix, and Boston (Brown et. al., 1983; Rosnow et al., 1985; Schutt and Garrett, 1986). Unless a woman has a very severe mental disorder, it is probably easier for her to remain within the family network or, failing that, to find some kind of housing. Therefore, it may be the more dysfunctional, alienated, or disorganized woman who becomes homeless.

Women who are heads of households and who are homeless with their children represent a segment of the homeless population that is growing both in numbers and in proportion. These women have a markedly different psychiatric profile than individual homeless adult women. Bassuk and colleagues (1986) have examined homeless mothers with families in approximately two-thirds of the family shelters in Massachusetts. Substance abuse was relatively rare, but this may be an underestimate since the study was limited to shelters and did not include hotels and motels. Of the mothers, 3 percent were schizophrenic, major affective disorders were found in 10 percent, and personality disorders were diagnosed in 71 percent.³ The children manifested considerable emotional and intellectual impairment. The authors stressed that measures to help such families, if they are successful, must attend to these psychiatric issues.

Clinicians who work with homeless people in primary health care clinics confirm the high frequency of psychiatric disorders in their patients. Brickner and coworkers (1985), in listing the common health problems encountered in a primary care program serving homeless people in shelters in New York City, placed alcoholism and then psychiatric disorders at the top of the list. The Health Care for the Homeless program's service data, pooled from sites in major cities across the country, show that, whatever the presenting problem, the primary care practitioners also frequently observed a mental or emotional disorder. The authors estimated that 30 to 40 percent of patients have psychiatric disorders (Wright and Weber, 1987). Clinicians were twice as likely to record a psychiatric diagnosis in white patients than in black or Hispanic patients. There was also a higher prevalence of almost every category of physical illness in patients with psychiatric diagnoses than in those without.

Mental disorders are very frequent in homeless populations generally and among homeless people who seek health care. In both groups mental disorders are found that can be considered both a cause and a consequence of homelessness.

Clinical Problems in Providing Mental Health Care for the Homeless

The central problem for homeless people with mental illnesses is the lack of community-based treatment facilities and adequate housing. In addition, the special characteristics of this patient group present particular challenges for treatment. These patients often have already had negative experiences with mental health care, often in understaffed, underfunded institutions, and are determined not to accept further treatment. Some have had unpleasant adverse reactions to antipsychotic medications or remember having been abused in the mental health care system; some homeless people lack insight into the reality of their illness and their need for ongoing treatment, but others who are aware of their problems simply do not believe that they will receive appropriate treatment if they accept an offer of care. In most cases, they lack the support of friends or family, are suspicious of authority figures (including providers of treatment), and are slow to develop a trusting therapeutic relationship. As is the case with the homeless in general, their material resources and access to public support programs are extremely limited (see Chapter 4).

From the perspective of mental health service providers, homeless patients are often perceived as less desirable or less rewarding. They may be slow to accept a therapist's sincere efforts to help, but quick to express their negative feelings about the mental health service system. A therapist may be frustrated by failures of homeless patients to keep appointments; and clinics may be unwelcoming to dirty, disheveled, or disorganized patients who frighten away others.

The treatment and rehabilitation of a severely mentally ill homeless person requires the marshaling of major financial and professional resources. Treatment requires enormous patience; considerable clinical skill; and the capacity to mobilize an array of treatment, residential, and rehabilitation resources to meet the needs of a particular patient (Breakey, 1987).

Although ambulatory treatment for mentally ill patients is preferred in most cases, hospital admission may be necessary for treating some patients with severely distressing and disabling symptoms, or for the protection of others if a person is violent. Psychiatric treatment providers are frequently frustrated in their efforts to help the most severely disturbed because of the lack of access to inpatient treatment facilities. The committee received reports from several cities and states that stated that because the supply of psychiatric beds is limited, some poor patients have great difficulty gaining access to voluntary inpatient care; occasionally there may even be a waiting period of several days at a public hospital for emergency involuntary psychiatric admission.

Consent for outpatient or inpatient treatment often can be obtained from a homeless patient relatively easily. For hospital care, voluntary admission is greatly preferred over involuntary commitment and facilitates the development of a constructive doctor—patient relationship. When a patient is unwilling to accept treatment but is clearly dangerous to himor herself or others, civil commitment procedures are available. However, problems arise when a patient is ill and behaves in a manner that is self-jeopardizing or is offensive, embarrassing, or frightening to others. Because these people are not unequivocally dangerous to themselves or others, they cannot be involuntarily committed.

Another problem confronting clinicians is a person who is neither offensive not dangerous but who is resistant to treatment because of delusions arising from the mental illness itself. Mental health workers may believe that medication and supportive care could substantially help a mentally ill person cope, but the patient is legally entitled to refuse treatment.

Mentally ill homeless people have attracted considerable attention from the news media. One of a series of articles in The New York Times in late 1986 called attention to a successful new approach to outreach, treatment, and rehabilitation (Goleman, 1986). The program was described by the journalist as "a partnership between modern psychiatry and older humanitarian traditions." Its various elements—persistent outreach, medication, professionally supervised supportive housing, disability benefits, preparation for employment, and so forth—certainly seem to have helped "Timothy":

A Year on the Road Back

Jan. 23, 1985. Mental health workers find Timothy huddled in a pile of garbage in a stairwell on West 68th Street. He says the Mafia is after him and laughs oddly. From his confused account, it appears that he has been hiding in garbage for at least two months. He resists efforts to move him to a shelter, preferring the stairwell.

March 4. After several false starts, Timothy is finally brought into the office of Project Reachout for treatment. One drug, then another is tried. He is put on a regimen of Prolixin, an anti-psychotic drug that diminishes agitation and delusions.

March 11. Meanwhile, a psychiatrist records improvement in Timothy's mental condition; his thinking is clearer, he is more alert, feeling better about himself. The patient starts to take showers.

March 19. For the first time, Timothy expresses interest in washing his clothes.

April 10. He agrees to leave the stairwell behind, accepting a tiny hotel room from the project.

May 19. He starts working in the kitchen at Fountain House, an organization that helps chronic schizophrenics to take part in society again.

Sept. 10. After project workers apply in his behalf, he receives his first Government disability payment.

Sept. 25. He begins work as a messenger at Manufacturers Hanover Trust bank offices. A project worker initially accompanies him to take over if he fails.

Oct. 29. He moves into a room of his own at the St. Francis Residence. There is a cable hookup for the color television he hopes to get. The room is newly painted. On the floor by the closet is a blue plastic bucket containing three pairs of filthy shoes and six umbrellas, mementos of his street days.

Alcoholism and Alcohol Abuse

In whatever setting homeless adults are studied, alcoholism is the most frequent single disorder diagnosed. (The exception is women who are homeless with their families.) Severe and intractable alcohol disorders have historically been thought to be especially prevalent among homeless people. Early accounts often attributed the high frequency of alcohol problems among homeless men to their inherent shiftlessness and failure to obtain gainful employment. Anderson (1923) asserted that "practically all homeless men drink when liquor is available. The only sober moments for many hoboes and tramps are when they are without funds." In 1946, Straus used prevalence figures from the studies that were available then and estimated that, overall, 80 percent of homeless men could be considered alcoholic. More recent studies, however, suggest that this perception may be stereotypical rather than real. It is now estimated that approximately 25 to 40 percent of homeless men suffer from serious alcohol problems, and that this level has been reasonably consistent over time (Mulkern and Spence, 1984; Stark, 1987; Schutt and Garrett, in press). This is nevertheless a high figure when compared with that in the general population, in which the most frequently reported figures are 11 to 15 percent for men and 2 to 4 percent for women. The prevalence of alcoholism for domiciled adults in the epidemiological catchment areas studied by NIMH is 12 percent for men and 2 percent for women (J. E. Heizer, Department of Psychiatry, Washington University School of Medicine, personal communication, 1988).

The emergence of a new homeless population further calls into question the meaning of previous findings. Current descriptive studies reveal a population that is younger and more heterogeneous than skid row populations. It includes (1) higher proportions of women and minority group members, such as blacks and Hispanics; (2) alarming numbers of families with young children; and (3) an increased proportion of people with mental illnesses and histories of drug abuse. Despite these changes, serious alcohol problems are common among homeless adults and remain important in understanding this population (see Table 3-9).

TABLE 3-9

Prevalence of Alcohol Abuse and Alcoholism Among the Homeless (in percent), 1982-1987.

Current studies also document the fact that homeless people with alcohol problems are more often physically disabled than homeless people without such problems. As a consequence, they are more likely to use health care services (Fischer, 1987; Fisher and Breakey, 1987; Koegel and Burnam, 1987a,b). Wright and Weber (1987) have identified specific disorders that occur more frequently among homeless alcoholics than other homeless people; these include acute disorders, such as trauma, serious skin problems, and severe upper respiratory infections, along with chronic disorders such as cardiac disease, hypertension, and active tuberculosis.

What is the relationship between homelessness and serious alcohol problems? Observers have indicated that many homeless adult individuals who suffer from alcoholism and alcohol abuse are undomiciled to begin with because of their drinking. In a study of homeless male alcoholics in Baltimore, 59 percent reported that drinking caused them to become homeless (Fischer and Breakey, 1987). Others may have become "environmental alcohol abusers" (Shandler and Shipley, 1987), adapting to a homeless subculture that encourages drinking. For homeless individuals, drinking is often seen as the way to make it through a day (Morgan et al., 1985) or to forget failure (Wiseman, 1987).

In the past, many alcoholics lived in SRO housing. For example, a 1979 study of 2,110 SRO housing residents in New York City indicated that 25 percent were alcoholics (Kasnitz, 1984). With the decline in the number of SRO units nationwide (see Chapter 2), many alcoholic single men and women have become homeless. Today many of the country's emergency shelters will not accept anyone who has been drinking. Instead, many homeless alcohol abusers sleep on the streets.

In sum, the causal relationships between problems with alcohol and homelessness are complex, and precise knowledge of them may not be possible or even as therapeutically relevant as one might hope. There is a large general literature on treating people who have problems with alcohol, even though the scientific evaluation of treatment in this area is relatively recent. Effective approaches to this population might have to include several elements, for example, detoxification, convalescence, and entry into specialized alcohol-free living environments combined with supportive treatment programs.

Detoxification is the indispensable first step in treatment; access to detoxification needs to be widely and readily available. Experience suggests that many people entering detoxification facilities will progress no further, and that a small number of people account for the majority of admissions. It is not always possible, however, to predict who will progress further with treatment; a common clinical experience is that, after multiple short-term admissions, some people elect to continue, and eventually they achieve genuine gains.

In recent years disagreement has arisen over the optimal structure of detoxification programs. Traditionally, detoxification has been undertaken in an inpatient medical setting. More recently, nonmedical detoxification programs have arisen (see, e.g., Annis et al., 1976). The latter have attracted much attention because of their markedly lower cost and reportedly equivalent effectiveness (McGovern, 1983). Ideally, a mixture of both would be available. There is little doubt that many people seeking detoxification can be handled in a nonmedical program (Shaw et al., 1981). However, withdrawal from alcohol in people with serious concurrent medical or psychiatric disorders is best undertaken in a hospital setting; many homeless people fall into this category.

After detoxification, some people are unable or unwilling to take advantage of the currently available rehabilitation alternatives, which at present require entry into a specialized alcohol treatment system. Some of these difficulties could be resolved if there were an intermediate stage in the treatment process between detoxification and specialized treatment (Blumberg et al., 1973; Shandler and Shipley, 1987). The goal of such a convalescent stage would be to facilitate complete recovery from the physical and mental ravages of the individual's last period of alcohol intake. A safe setting, perhaps best outside of, but closely connected to, a medical facility, could provide protection, adequate nutrition, rest, and an opportunity to assess the future realistically. Extended medical and psychiatric evaluations, which are problematic in detoxification settings, could be performed, and consequent therapeutic measures could be proposed.

Specialized treatment and active rehabilitation for alcohol-related problems are complex (a forthcoming Institute of Medicine study will explore this subject in detail). Some homeless people with alcohol-related problems may eventually enroll in specialized treatment. However, access to such specialized treatment is far from universal, and the shortage of facilities is serious. Furthermore, there is an extreme shortage of the specialized housing arrangements that are needed to support rehabilitation efforts. Residential opportunities are essential to enable the alcoholic homeless person to get away from the streets, where inducements to resume drinking are ever present.

Illnesses Associated with Abuse of Drugs Other Than Alcohol

There are few concrete data describing the extent of drug abuse among homeless individuals. Most studies about the homeless combine alcohol and drug abuse together under the heading of substance abuse. Those that separate the two provide some minimal information about illicit drug use. Estimates of homeless individual adults with drug problems range from a low of 10 percent reported by users of Johnson-Pew clinics nationwide (Wright and Weber, 1987) to 33.5 percent for individuals living in shelters and on the streets in Boston (Mulkern and Spence, 1984) (see Table 3-10).

TABLE 3-10

Illicit Drug Use Among Homeless Individuals.

Data from the Johnson-Pew HCH projects in 16 cities show that age, ethnicity, and drug abuse are correlated (Wright, 1987). The strongest correlate of drug abuse is age. As with the general domiciled population, rates of illicit drug abuse are highest among younger HCH clients and fall off with increasing age, especially after the age of 50. This is almost the opposite of alcohol abuse, which is found to be least prevalent among younger homeless people.

One of the problems associated with drug abuse is AIDS and AIDS related complex (AIDS/ARC in Tables 3-5 and 3-6). Whether this is more commonly encountered among homeless people who abuse drugs compared with the remainder of the drug-abusing population is not clear. Nevertheless, as the clinical syndromes associated with AIDS increase in the general population, especially among those who abuse parenteral drugs, it will be an increasing problem among the homeless as well.

Working from a series of assumptions and data from the 16 cities of the Johnson-Pew HCH projects, Wright and Weber estimated the prevalence of AIDS and AIDS-related complex at about 185 per 100,000 homeless persons in those cities. Meaningful comparisons to the rates of AIDS in each of those urban centers are not available, but one reference point is the rate for the U.S. population as a whole, viz. approximately 144 per 100,000 in 1986 (Wright and Weber, 1987).

Other illnesses more commonly encountered in patients who abuse parenteral drugs are hepatitis, skin infections, abscesses, thrombophlebitis, bacterial endocarditis, and tuberculosis. Other, more exotic infections that are not frequent in the United States are more common among drug abusers, such as malaria, which can be transmitted among patients who share needles. SADRI specially analyzed its main data base, which consisted of all clients with two or more visits who abused drugs, and found that some disorders were more common among drug abusers than among non-drug abusers. To some extent, however, the differences could be ascribed to various demographic characteristics, specifically, age and the presence of other disorders such as alcohol abuse or mental illness. Using this series of multivariant analyses, which controlled statistically for age, sex, ethnicity, and family status and for alcohol abuse and mental illness, the following disorders were found more commonly among homeless people who were drug abusers: AIDS, liver disease, cardiac disease, peripheral venous stasis disease, and chronic disorders such as diabetes and diseases of the liver and genitourinary tract. Although the exact relationship between homelessness and drug abuse and these illnesses is unclear, most of the findings are not surprising. AIDS and liver disease, for example, are associated with an increased frequency of hepatitis exposure among drug abusers.

Comorbidity

Finally, a point must be made about the comorbidity caused by mental illness, alcoholism and alcohol abuse, and illicit drug abuse. There is a growing concern among those who work with homeless people about clients with dual and multiple diagnoses (further exacerbated by a higher prevalence of many acute and chronic physical illnesses). For example, the HCH data point to correlations among drug abuse, alcohol abuse, and mental illness. Among drug abusers, 42 percent of the men and 41 percent of the women who visited HCH projects and gave evidence of that diagnosis could also be classified as mentally ill; 59 percent of the male clients and 46 percent of the female clients who abused drugs also evidenced a problem with alcohol (Wright, 1987). In another recent study drawn from a broad geographic base, the Veterans Administration Homeless Chronically Mentally Ill program reported that of the homeless for whom evaluations were performed, 32 percent had combined diagnoses of alcohol and drug abuse. Sixty-four percent had been hospitalized for any treatment for mental illness, alcoholism, or drug abuse. Because this latter figure is less than the sum of the prevalence rates for homeless veterans seen for each diagnosis (33 percent reported being hospitalized for psychiatric illness, 44 percent for alcoholism, and 14 percent for drug abuse), it appears that many of these hospitalizations were for dual or multiple diagnoses (Rosenheck et al., 1987).

There are two major problems that relate specifically to homeless people with multiple diagnoses. During the site visits, it was repeatedly emphasized to the committee members by those who work with the homeless that homeless people with dual and multiple diagnoses are among the most difficult to entice into treatment. Second, when outreach efforts are successful, there often are no appropriate programs into which such homeless people can be enrolled. Each separate diagnosis correlates to a specific treatment modality and treatment system. These programs frequently exclude those with secondary and tertiary diagnoses of other illnesses. It is rare to find programs that will address a combination of diagnoses on other than the most episodic of terms.