Homelessness and Mental Health
Homelessness and Mental Health
Homelessness and Mental Health
PETER J. TYRER
Professor of Community Psychiatry, St Mary's Hospital Medical School, London
Also in this series
IAN
R. H. FALLOON
M.
PHELAN,
G. STRATHDEE
AND
Homelessness and
Mental Health
edited by
DINESH BHUGRA
Institute of Psychiatry
London, UK
CAMBRIDGE
UNIVERSITY PRESS
Dedicated to
Suman and Rajesh and their families
Contents
List of contributors
Preface
Part I
1
xii
xv
I N T R O D U C T I O N AND SPECIAL G R O U P S
Introduction
DINESH BHUGRA
11
P H I L I P TIMMS
Models of homelessness
26
A L A N M G N A U G H T AND D I N E S H BHUGRA
41
DINESH BHUGRA
Homeless women
59
E. J A N E M A R S H A L L
78
PHILIP JOSEPH
Part II
7
SERVICES
Services for the mentally ill homeless
99
E. J A N E M A R S H A L L AND D I N E S H BHUGRA
IX
Contents
8
110
BREAKEY
133
MAX MARSHALL
10
150
C A R L I. C O H E N AND M A U R E E N C R A N E
11
170
Part III I N T E R N A T I O N A L P E R S P E C T I V E
12
European perspectives
Introduction to European chapter
187
MAX MARSHALL
Homelessness in Denmark
189
P R E B E N B R A N D T AND
POVL M U N K - J 0 R G E N S E N
197
W U L F R 6 S S L E R AND
HANS JOACHIM SALIZE
209
JOSEPH FERNANDEZ
13
14
230
BAGHRAGH
244
Contents
xi
267
DAVID KINGDON
16
280
MAX MARSHALL
17
297
DINESH BHUGRA
Index
307
Contributors
Contributors
xiii
D R J O S E P H FERNANDEZ
xiv
Contributors
Preface
'I had a home once - I had once a husband I am a widow, poor and broken-hearted!'
Loud blew the wind, unheard was her complaining.
On drove the chariot.
The Widow
Robert Southey 1774-1843
xvi
Preface
PART I
INTRODUCTION AND
SPECIAL GROUPS
1
Introduction
DlNESH BHUGRA
4-
D. BHUGRA
Introduction
D.BHUGRA
Introduction
D. BHUGRA
Introduction
10
D. BHUGRA
2
Homelessness and mental illness:
a brief history
PHILIP TIMMS
12
P. TIMMS
13
14
P. TIMMS
15
16
P. TIMMS
The general feeling that the workhouse was not a good place for the
mad resulted in 1875 in a weekly grant to all unions in England and
Wales from the Poor Law Board. Four shillings per person per week
was to be paid towards the extra cost of maintaining pauper lunatics
in county asylums (Cochrane, 1988). As the entire cost would
otherwise fall upon the local parish rates, this was a powerful
inducement and could result in the saving of 60% of a local
metropolitan union's expenditure on such people. However, in 1889
14% (11827) of pauper lunatics in England and Wales outside
London were still confined in workhouses, 8% at home and 78% in
county asylums. However, in London only 2% (275) were accommodated in workhouses, 2% at home and 96% in London City
Council (LCC) asylums. This can be related to the LCC's vigorous
programme of asylum-building, prompted by the expense of paying
for placements outside the LCC area. So, outside London at least, the
workhouse continued to be a substantial provider of shelter for the
mentally ill up until the First World War.
From then until the Second World War very little seems to have
been written on the subject. Even George Orwell does not seem to
have noticed any madness in his perambulations through the world of
the destitute in the London and Paris of the 1930s. To be fair, he was
at pains to stress the ordinariness of the men who found themselves
17
18
P. TIMMS
19
20
P. TIMMS
21
22
P. TIMMS
Schizophrenia Alcoholism
(%)
(%)
Affective
disorder (%)
Personality
disorder (%)
24
25
N/K
N/K
20
N/K
N/K
66
32
15
18
N/K
5
8
18
51
-25
25
-5
17
Conclusion
Homeless people, and thus the homeless mentally ill, have been a
constant feature of the English social landscape since the fourteenth
century. The mechanisms for dealing with the problem have been
moulded by the changing tempers of the times. Throughout the
centuries, societal attitudes towards the homeless and the homeless
References
Allderidge, P. (1979). Hospitals, madhouses and asylums: cycles in the
care of the insane. British Journal of Psychiatry, 134, 321-33.
Anderson, N. (1922). The Hobo: The Sociology of the Homeless Man.
Chicago: University of Chicago Press.
23
24
P. TIMMS
Wagnalls.
Cochrane, C. (1988). The Asylum and its Psychiatry. Routledge: London.
Coid, J. W. (1988). Mentally abnormal prisoners on remand. I.
Rejected or accepted by the NHS? British Medical Journal, 296,
1779-82.
Craig, T. & Timms, P. W. (1992). Out of the wards and onto the
streets? Deinstitutionalisation and homelessness in Britain. Journal of
Mental Health, 1, 265-75.
Crossley, B. & Denmark, J. C. (1969). Community care - a study of the
psychiatric morbidity of a Salvation Army hostel. British Journal of
Sociology, 20, 443-9.
De Schweinitz, K. (1943). England's Road to Social Security. Pittsburgh:
University of Pennsylvania Press.
Edwards, G., Williamson, V., Hawker, A., Hensman, C. & Postsyan, S.
(1968). Census of a reception centre. British Journal of Psychiatry,
114, 1031-9.
Faris, R. E. & Dunham, H. W. (1959). Mental Disorders in Urban Areas:
an Ecological Study of Schizophrenia and other Psychoses. Chicago:
25
Orwell, G. (1933). Down and Out in Paris and London. London: Penguin.
Parry-Jones, W. L. (1972). The Trade in Lunacy. London: Routledge.
Paul, G. O., Sir (1812). Observations on the Subject of Lunatic Asylums.
Gloucester: private publication.
Porter, R. (1989). Mind-forged Manacles: A History of Madness in England
from the Restoration to the Regency. London: Penguin.
Power, E. & Tawney, R. H. (eds.) (1924). Tudor Economic Documents, vol.
II.
Priest, R. G. (1976). The homeless person and the psychiatric services:
an Edinburgh survey. British Journal of Psychiatry, 128, 128-36.
SHIL (Single Homelessness in London) (1986). A Report by the Single
Homelessness in London Working Party. London: SHIL.
Southern, R. W. (1970). Western Society and the Church in the Middle Ages.
Harmondsworth: Penguin.
The Resettlement Units Executive Agency. (1991). Annual Report and
Financial Statement 1990/91. London: HMSO.
Tidmarsh, D. (1972). Services for the destitute: Camberwell reception
centre. In J. Wing & A. M. Hailey (eds.), Evaluating a Community
Psychiatric Service. The Camberwell Register, 1964-1971, pp. 73-6.
Oxford: Oxford University Press.
Tidmarsh, D. & Wood, S. (1986). Report to DHSS on Research at
Gamberwell Reception Centre. London: Institute of Psychiatry.
(Unpublished).
Trevelyan, G. M. (1986). English Social History. London: Penguin.
Vexliard, A. (1953). Le Clochard: Etude de Psycholgie Sociale. Paris: Desclee
de Brouwer.
Webb, S. & Webb, B. (1929). English Poor Law History: Part 2: The Last
Hundred Tears. London: Longmans, Green & Co.
Whiteley, J. S. (1955). Down and out in London: mental illness in the
lower social groups. Lancet, 2, 608-10.
Wilmanns, K. (1906). Zur psychopathologie des landstreichers: eine
klinishe studie. Leipzig: Johann Ambrosius Barth.
3
Models of homelessness
ALAN M G N A U G H T AND DINESH BHUGRA
Introduction
There is no single definition of homelessness accepted by all who use
this term. Homelessness exists throughout this and other nations as
will become extremely clear in the accounts in this book, and the
homeless population is diverse.
In the UK, the National Assistance Act of 1948 requires every local
authority to provide 'temporary accommodation for persons who are
in urgent need thereof, being need arising in circumstances that
would not reasonably have been foreseen'. Hence, local authorities
have a statutory obligation to provide accommodation to individuals
who are roofless, and have not placed themselves in this state. In
practical terms, this definition of homelessness is probably the most
important one for individuals living in the UK. When these conditions are fulfilled, local authorities have a statutory obligation to act
and provide shelter to the homeless person.
Whether individuals have been 'blameworthy' in becoming homeless, or not, remains an arbitrary decision made by local authorities.
Because resources are scarce, a narrow working definition of homelessness tends to be used. Local authorities are more likely to embrace
the needs of homeless families, rather than the single homeless, hence
the latter group are more likely to be excluded, under the terms of the
Act.
Voluntary agencies working with homeless people have tended to
use a far broader definition of homelessness. The charity Shelter
working for the homeless, for instance, extends the definition of
homelessness to those living in very bad housing conditions. Thus
people living in very poor quality housing, or sleeping over with
26
Models of homelessness
27
28
Classification
Within this diversity of homeless people there are, however, subgroups of homeless individuals who share characteristics and may
also share certain needs. Classifying homeless people is of use if
identification of a subgroup of homeless people leads to greater
understanding of how individuals in that group became homeless and
how they can be helped. Classification can be based on aetiology
(how individuals became homeless), on current impairment or on
socio-demographic factors. Classification of any group of people
should, ideally, include all members of the group, in discrete and
readily recognizable subgroups of individuals.
Leach (1979) proposed that homeless people should be divided
into 'intrinsics' and 'extrinics'. 'Intrinsics' were homeless people
whose mental or physical disability was the cause of their homelessness, and had therefore predicted their homelessness. 'Extrinsics', by
contrast, were homeless because of situational factors (e.g. job loss,
poverty etc.). He stated that the two groups would require different
services. Their needs and priorities will differ as will their management.
Morse et al. (1991) proposed a system that focuses on current
psychiatric or physical impairment, regardless of whether such
impairment was the original cause of the homelessness, or not. They
demonstrated that both systems identified subgroups with different
needs, but that their system was superior at identifying groups of
individuals with high levels of service needs. Cohen (1994, and
Chapter 10) suggests that the characterizations of homelessness as a
trait rather than a state reflect the tensions between social justice and
public concepts as well as economic resources.
The UK Royal College of Psychiatrists Working Party Report on
Homelessness and Mental Illness (Bhugra, 1991) recommended that
three categories of homeless people be recognized: the single homeless, young single homeless and homeless families. The single homeless
have been the most extensively investigated group, to date. The
group consists of single men and women over 25 years of age. Wright
29
Models of homelessness
Social factors
Political economy
Housing policy
Unemployment
Geography
Duration and
nature
"
-Throwaways/Runaways
Ii
i
i
Homelessness U-Poverty
-Unemployment
t \
Individual
Old
Families
Figure 3.1. Pathways into homelessness and types and interaction of various factors.
(1989) argues that subgroups of homeless are very different and they
can be divided according to social origins, background characteristics
and needs. The portrait can be of a pathway into homelessness along
with uniqueness of each group. In addition he challenges the concepts
of deserving and undeserving homeless people. These remain moral
judgements. He classified homeless as homeless families, lone homeless (which were further subdivided into men, children and women),
and then those with special factors like disability, ethnicity, education, geography, criminality, extreme poverty and nature and
reasons for homelessness. In the study conducted by Wright & Weber
(1987) 16-26% belonged to homeless family groups and the rest were
defined as lone homeless. Furthermore, 6% were children or adolescents below 19 years of age, 20% were adult women and 74% were
adult men (see Figure 3.1 for pathways into homelessness and types of
homelessness). This is not a clear, linked pathway and various factors
influence it along the way. The vast majority of the single homeless
30
Models of homelessness
31
32
Models of homelessness
33
34
Models of homelessness
35
36
Models of homelessness
37
38
References
Augermeyer, M. C., Goldstein, J. M. & Kuhu, L. (1989). Gender
differences in schizophrenia: rehospitalization and community
survival. Psychological Medicine, 19, 365-82.
Bachrach, L. L. (1984). Interpreting research on the homeless mentally
ill: some caveats. Hospital and Community Psychiatry, 35, 91417.
Bardenstein, K. K. & McGlashan, T. H. (1990). Gender differences of
affective, schizoaffective and schizophrenic disorders: a review.
Schizophrenia Research, 3, 159-72.
Barons Court Report (1993). Barons Court Project - six month
monitoring report. Oct 1992 to March 1993. London: Barons Court
Project.
Bassuk, E., Rubin, L. & Laurant, A. (1986). Characteristics of sheltered
homeless families. American Journal of Public Health, 76, 1097-101.
Bhugra, D. (1991). Homelessness and Mental Illness. Working Party
Report, Royal College of Psychiatrists, London.
Bhugra, D. (1993). Unemployment, poverty and homelessness. In D.
Bhugra & J. Leff (eds.), Principles of Social Psychiatry, pp. 335-84.
Oxford: Blackwell Scientific Publications.
Burt, M. R. & Cohen, B. E. (1989). Differences among homeless single
women, women with children and single men. Social Problems, 36,
508-24.
Cohen, C. I. (1994). Down and out in New York and London: a cross
national comparison of homelessness. Hospital and Community
Psychiatry, 45, 769-76.
Cohen, C. I. & Thompson, K. S. (1992). Homeless mentally ill or
mentally ill homeless. American Journal of Psychiatry, 149, 816-23.
Dayson, D. (1993). The TAPS project: crime, vagrancy and
readmission. British Journal of Psychiatry, 162 (Supplement 19), 40-4.
Gelberg, L., Linn, S. & Leake, B. D. (1988). Mental health, alcohol and
drug use, and criminal history among homeless adults. American
Journal of Psychiatry, 145, 191-6.
Goering, P., Paduchak, D. & Durbin, J. (1990). Housing homeless
women: a consumer preference study. Hospital and Community
Psychiatry, 41, 790-4.
Herzberg, J. (1987). No fixed abode: a comparison of men and women
admitted to an east London psychiatric hospital. British Journal of
Psychiatry, 156, 621-7.
Hirsch, S. R. (1992). Services for the severe mentally ill: a planning
blight. Psychiatric Bulletin, 16, 673-5.
Koegel, P. & Burnam, M. A. (1988). Alcoholism among homeless adults
in the inner city of Los Angeles. Archives of General Psychiatry, 45,
1011-18.
Koegel, P., Burnam, M. A. & Farr, R. K. (1988). The prevalence of
specific psychiatric disorders among homeless individuals in the
inner city of Los Angeles. Archives of General Psychiatry, 45, 1085-92.
Models of homelessness
39
40
Introduction
The most alarming increase in the homeless population over the last
20 years has been the dramatic rise in the number of homeless families
with children. The homeless youth have appeared in increasing
numbers on our streets. In England, Shelter (a charity for the
homeless) estimated that 58 000 people were either sleeping rough or
were unauthorized tenants or squatters. Up to another 137 000 single
people were estimated to be living in hostels and lodgings in addition
to 317 000 insecure private tenants and 1.2 million 'hidden' homeless
(Burrows & Walentowicz, 1992). In the USA in 1986 it was reported,
following a survey of 25 cities, that 80% of the sample had reported an
increase in the number of homeless families with children (US
Conference of Mayors, 1986). A year later on average the numbers
had gone up by another one third (US Conference of Mayors, 1987).
Various changes in housing policies and a reduction in available
housing stock available at affordable rents along with low cost new
building are among some of the reasons for this increase (see Chapter
3). For homeless families stresses and coping strategies are different
from those experienced by the single homeless. In homeless families in
addition to individual stress, there is also the impact on the family
unit along with all of the interactions that happen between the unit
members. Within that unit, family homelessness can affect not only
emotional growth but behavioural and nutritional status of the
children.
The stereotypes of homeless youth running away towards the
bright lights does not hold up to close scrutiny. There is no typical
homeless family or runaway or homeless youth. These individuals
41
42
D. BHUGRA
Family
Poverty
Unemployment
Homeless
Shelters
Parents -+
Bed and
Breakfast
Children-
Other factors
Squatting
Living with friends
- Educational
deficits
Depression
Behaviour disturbances
Developmental delays
Emotional disturbances
- Repeated infections
Physical ill health
Poor nutritional status
Figure 4.1. Pathways into homelessness and consequences of homelessness on the
family.
come from a range of socio-economic classes, races, ages, backgrounds, and their needs, whether physical, social, economic or
health, also vary.
In this chapter the needs of homeless families, homeless children
and homeless youth will be addressed and suggestions for identification and management of problem areas will be made.
Homeless families
Over the past decade there has been a large increase in the number of
households in England accepted as officially homeless from 53 110 in
1978 to 145 800 in 1990 (Faculty of Public Health Medicine). The
number of families in London accepted as homeless under the terms of
the 1985 Housing Act increased by nearly threefold.
Figure 4.1 outlines the sequelae of homelessness in families. This
figure is not meant to be comprehensive but indicates many of the
problems of homeless families that need to be assessed multi-axially.
Homeless families in the UK can be divided into those who are
43
44
D. BHUGRA
unit. These mothers had two or more children, were less well
educated and had limited work experience. The last category
included women who had been abused, were younger (in their early
20's), usually with one child often an infant. They may have had a
history of fostering and may have run away from this placement often
because of sexual abuse. This group was described as having learnt
subsistence prostitution. These mothers were also more likely to need
support and rehabilitative services attached to specialized housing
alternatives (Bassuk et al., 1986). Making facilities for child care
available along with the benefits of voluntary help and voluntary
organizations are important factors in planning their management.
Bassuk et al.'s (1986) sample from Massachusetts included 80
homeless women and 151 children drawn from 14 family shelters.
Women were young and over two thirds came from broken homes.
They had poor work records, few supports and long histories of
residential instability. Nearly three quarters (71%) were given a
diagnosis of personality disorder. Boxhill & Beaty (1990), in a study of
40 homeless women and their children, observed that these mothers
and children were building their relationships in an open personal
space. The authors reported six themes to emerge from their study.
These were: an intense desire to demonstrate internalized values as a
way of asserting self; questioning the certainty of anything and
ambiguity of every thing; conflict over the need for attention and the
experienced demand for independence; public mothering; unravelling of the mother role; external control being placed on this maternal
role. Little wonder that an internalized confusion may arise within
mothers and their relationships. Smith & North (1994) studied 300
homeless women (90% of whom were mothers) in St Louis from night
shelters and day centres. Using the Diagnostic Interview Schedule/
Homeless Supplement, DSM-III R diagnoses were reached. Two
third (67%) of the women had children under the age of 16 in their
physical custody in these shelters. Fifty women (17%) did not have
physical custody of any of their children under 16. Unlike previous
studies this sample was predominantly (76%) African American and
mothers with children were more likely to be unemployed. The
authors suggest that in their sample, women with children had
greater social vulnerabilities (like unemployment and dependent
children) and fewer personal vulnerabilities (substance abuse and
other psychiatric problems). Among those without children nearly
half had a non-substance Axis 1 life-time psychiatric disorder and
45
46
D. BHUGRA
under the age of 5. In studies from the USA it has been estimated that
on any given night there are 6800 children and youths aged 16 or
under who belong to homeless families (GAO, 1989). In addition a
further 300000 children may be counted as homeless per year.
Furthermore, another 186000 children may be doubled up in
housing shared with relatives or friends. The GAO study reported
that 52% of homeless children in their survey were aged 5 or younger
- nearly two thirds of children aged 16 or under seen by the National
Health Care for the Homeless projects were aged 1 to 5. Most of the
families in the shelter belonged to ethnic minorities (Bassuk, 1987,
Miller & Lin, 1988). As noted earlier, Victor (1992) reported from
London that being female and young was related to being housed in
bed and breakfast hotels. Of the 522 subjects, 207 (55%) had between
one and seven dependent children under the age of 16, most parents
having a single child of school age. Solarz (1992) emphasizes that a
separate group of families and children exist who may survive by
living rough. There is no doubt that in the studies reported so far the
children have borne the brunt of physical and mental health
problems.
Bassuk's (1986) work referred to earlier, revealed that only
women's relationships with men were characterized by instability,
conflict and violence. The children from this study had been
administered the Denver Developmental Screening Test to test
developmental milestones in four areas: language, gross motor skills,
fine motor co-ordination and personal and social development. The
children interviewed in the series revealed very high (47% of
pre-schoolers) rates of at least one developmental delay. One third
(33%) had lags in two or more areas. The school-age children
suffered from serious emotional problems. On the children's Depression Inventory and the children's Manifest Anxiety Scale about half
scored high enough to warrant fuller psychiatric evaluation. Almost
all had thought of committing suicide. Most of the older children
were doing poorly in school - one quarter were in a special class and
nearly half (43%) had to repeat a year. In addition some children
were ashamed of their social status of being homeless and kept it
hidden from schoolmates. Furthermore, this group did not generally
contact helping agencies either. Only 14% of children were involved
in day care. The immediate cause of homelessness was housing
related.
Whitman et al. (1990) have also reported language development
47
48
D. BHUGRA
49
50
D. BHUGRA
51
52
D. BHUGRA
53
wandering around from one address to the next) the author collected
basic information on 212 young people and interviewed 86 at length.
Boys were three times more likely to be homeless and foreign youth
formed one third of the sample. Of the 90% that were jobless about
one third had been unemployed for over a year. Van der Pleog (1989)
characterized these along five dimensions:
1. Negativefamily backgrounds. Nearly three quarters (73%) came from
families where parents had been divorced and 70% rarely kept in
touch with their fathers.
2. History ofprofessional help. Of the sample, 12% had been involved
with seven or more helping organizations, 79% had been in
residential institutions and 21% had been with foster families.
3. Negative experiences at school. Of the youth, 22% were persistent
truants and 70% had changed schools more than once, while 40%
had repeated one or more classes and 10% had changed school
eight times.
4. Few friends. Defining firm friendship as lasting at least one year
only 15% had friends, and 11% had no friends at all. This may
reflect the mobility as well as social skills and poor access to peer
groups for forming friendships.
5. Low self perception. These youths had a positive opinion of themselves and the majority saw themselves as adult, clever, self assured
and cheerful. Van der Ploeg argued that this was a facade and that
these youths had an inadequate self-perception and a poor
perception of their situation. Over 50% believed that their
destinies were determined by factors outside their control. Wiggans (1989b) criticized Van der Ploeg's model on the grounds that
Van der Ploeg had failed to outline the causal model that had been
the original intention and, in addition, the reality of the focus was
on pathology and weakness rather than the individual's strength.
54
D. BHUGRA
adolescents
Conclusions
The underlying causes of homelessness are varied and for the three
groups of homeless families, children and adolescents the individual
needs must be determined, thus allowing prioritization of care and an
early response to reduce the period of uncertainty, anguish and
homelessness. On the one hand homeless children suffer from a
multitude of problems and the educational aspects of their life-style
need to be highlighted. Homeless youth, on the other hand, are more
likely to be the products of broken homes, are as likely to run away as
to be thrown out and are likely to end up in prostitution with its
inherent risks. It is possible that all these factors are interdependent.
The situation is complex but identification of the special needs of
homeless families, youth and children, followed by a clear assessment
55
56
D. BHUGRA
References
Abrahams, C. & Mungall, R. (1992). Runaways: Exploding the Myths.
London: National Children's Home.
Bassuk, E. L. (1987). The feminization of homelessness: families in
Boston shelters. American Journal of Social Psychiatry, 7, 19-23.
Bassuk, E. L., Rubin, L. & Lauriat, A. S. (1984). Is homelessness a
mental health problem? American Journal of Psychiatry, 141, 1546-50.
Bassuk, E. L., Rubin, L. & Lauriat, A. S. (1986). Characteristics of
sheltered homeless families. American Journal of Public Health, 76,
1097-101.
Benjamin, M. (1985). Juvenile Prostitution: A Portrait of the (Life\ Toronto:
Ministry of Community & Social Services.
Bhugra, D. (1993). Unemployment, poverty and homelessness. In D.
Bhugra and J. Leff (eds.) Principles of Social Psychiatry, pp. 355-84.
Oxford: Blackwell Scientific Publications.
Boxill, N. A. & Beaty, A. L. (1990). Mother/child interactions among
homeless women and their children in a public night shelter in
Atlanta, Georgia. Child and Youth Services, 14, 49-64.
Burrows, L. & Walentowicz, P. (1992). Homes Cost Less Than
Homelessness. London: Shelter.
Covenant House (1985). The Covenant Experience. Toronto: Covenant
House.
CSA (Council on Scientific Affairs) (1989). Health care needs of
homeless and runaway youths. Journal of the American Medical
Association, 262, 1358-61.
Department of the Environment (1982). Single and Homeless. London:
HMSO.
Focus Point (1987). Towards a Resettlement Strategy and Annual Report.
Dublin: Focus Point.
Feitel, B., Margetson, N., Chamas, J. & Lipman, C. (1992).
Psychosocial background and behavioural and emotional disorders
of homeless and runaway youth. Hospital and Community Psychiatry,
43, 155-9.
GAO (General Accounting Office) (1989). Children and Youths.
Washington, DC: General Accounting Office.
Goldberg, D. (1978). General Health Questionnaire. Windsor:
NFER-Nelson.
Greenblatt, M. & Robertson, M. J. (1993). Life styles, adaptive
strategies and sexual behaviours of homeless adolescents. Hospital
and Community Psychiatry, 44, 1177-80.
57
58
D. BHUGRA
5
Homeless women
E. JANE MARSHALL
Introduction
The proportion of women amongst the adult homeless population
increased throughout the 1970s and 1980s and is currently estimated
at between 10% and 25%. Women with children are now the fastest
growing segment of the homeless population in the USA (Smith &
North, 1994). Information about homeless women is limited to
studies that have been carried out in accessible settings such as hostels,
shelters and day centres, and cannot, therefore, be generalized. In
studies where homeless men and women have been interviewed, the
numbers of women are small and comparisons between the sexes have
not been made (Arce et aL, 1983; Bassuk et al., 1984; Fischer et al.>
1986; Kroll^a/,,1986).
The current situation and the specific problems experienced by
homeless women in Britain and the USA can be better understood in
the light of historical evidence.
60
E. J. MARSHALL
Homeless women
61
prices beyond what the women could afford or would refuse accommodation (Higgs & Hay ward, 1910). Homeless women also eschewed the casual wards because of the punitive treatment meted out
and only used them as a last resort (Higgs & Hayward, 1910).
During the interwar period, homeless women could be divided into
one group that was temporarily stranded and two larger groups that
were permanently homeless (Menzies, 1927). The temporarily homeless included young women who had run away from home or who had
drifted up from the country in search of work, and older women with
a history of'drink and drug addiction, pilfering, evil temper or some
mental twist to account for their plight' (Menzies, 1927). These
temporarily stranded women usually found accommodation in facilities run by charities. The permanently homeless consisted of
(1) unskilled workers, vagrants and down-and-outs and (2) prostitutes. The unskilled workers and prostitutes largely stayed in the
common lodging houses, which remained a major source of provision
for homeless women. The casual wards were a less important source of
accommodation for homeless women and were used mainly by older
women, vagrants and down-and-outs.
In 1925 there were 17 310 places in London County Council
licensed lodging houses, but only 1630 (9.42%) were available for
women and conditions were worse for women than for men (Chesterton, 1926). This was another ploy to turn women away based on the
perception that they were more difficult to manage! For instance,
there was a general lack of washing facilities in womens' lodging
houses, whereas in mens' establishments there were baths. A 1930
census reported that the casual ward population on one night in
London and several other provincial poor law unions was composed
of 2472 men, but only 110 women (Watson & Austerberry, 1986). Of
these women, 56 were aged between 40 and 60 years and a further 25
were over 60 years.
Mrs Cecil Chesterton lived as a destitute woman on the streets of
London for a short period of time and recorded her experiences in a
book entitled 'In Darkest London' (Chesterton, 1926). Her vivid
description of what it was like to stay in Salvation Army and other
shelters, in licensed and unlicensed lodging houses, in doss houses, the
casual ward and also on the streets is still relevant. As a homeless
woman, her life came to be 'governed by an obsession for food and
sleep' and she understood how the destitute experience 'effaces
individuality, weighs down the will, clogs the instinct to do battle
62
E. J. MARSHALL
which is the heritage of man'. 'One instinct only remains vital, apart
from the desire for food and shelter, and that is the passionate
determination not to be trapped into an institution', by which she
meant the casual ward. She concluded that women became homeless
because of'poverty, . . . the shortage of housing ..., illness, bad luck,
increase of rent', not because of 'moral delinquency' and that the
accommodation provided for the destitute woman was inadequate
and generally of a low standard. When she returned home she set up
the Cecil Houses (Inc.) Public Lodging Houses Fund and later houses
for homeless single women that were financed by voluntary donations.
After the Second World War the only statutory provision for single
homeless women was the casual ward. These were renamed 'reception centres' and were administered by the new National Assistance
Board. Conditions were less punitive and the aim was to help the
single homeless to 'settle down'. That homeless women did not use the
reception centres was confirmed in a 1952 census, which reported that
numbers rarely exceeded 100 women on any one night (Watson &
Austerbery, 1986). Common lodging houses were declining in number and importance and one study reported that most women in
Glasgow's lodging houses were over 60 years of age and widowed
(Laidlaw, 1956). Privately rented accommodation was the most
important option for single people.
In the 1950s, legislation decontrolling rents and property development, particularly in cities, led to an increase in the cost of privately
rented accommodation and the closure of many common lodging
houses and hostels.
A census of single homeless accommodation carried out by the
National Assistance Board (1966) reported that 29 798 men and 1905
women were living in reception centres, common lodging houses or
hotels, or sleeping rough. The report did not consider why there were
so few homeless women, compared with men, nor did it assess their
characteristics. The 'concealed homeless' were ignored. This census
was updated in 1972 (Digby, 1976) when it was estimated that 23 300
single homeless men and 2200 single homeless women were living in
hostels and lodging house accommodation nationally. But during the
intervening years bed numbers for women had decreased, so that by
1972 the voluntary sector (e.g. Salvation and Church Armies)
provided almost 75% of the total beds for women and the private
sector provided only 14%. There is evidence that many of the women
Homeless women
63
64
E. J. MARSHALL
Homeless women
65
Table 5.1. Age of homeless women compared with homeless men: British
studies
Study
Men:Women
Men
Women
Digby (1976)
Drake*/*/. (1982)
Herzberg (1987)a
Fernandez (1984)b
Marshall & Reed (1992)a
Adams (1991)
Scott (1991)
James (1991)a
1821:172
398:122
62:48
15:28
0:70
0:64
0:46
0:43
43%<50
29%<30
40.5
40.5
51%<50
62%<30
36.3
37.0
52.1
45%<44
70%<40
32.7
a
b
Mean age.
Median age.
66
E. J. MARSHALL
MenrWomen
Men
83%: 17%
150:43
0:80
298:239
75:25
0:300
Women
years. In Scott's (1991) study, 70% of the sample was under 40 years,
whereas the mean age in Marshall & Reed's (1992) sample was 52.1
years.
Research from the USA corroborates the fact that homeless women
are younger than homeless men (Table 5.2). The homeless women in
Breakey et a/.'s (1989) study were even younger than samples from
Great Britain and Ireland and Lipton et a/.'s (1983) hospital sample
was also younger. Homeless mothers are the youngest group overall
with a mean age in the late twenties (Bassuk et al., 1986; Smith &
North, 1994).
Homeless women are more likely than homeless men to have stayed
on at school, to have obtained job training and to be employed
(Digby, 1976; Drake et al., 1982). They are also more likely to have
maintained contact with their family, to have married and had
children (Digby, 1976; Herzberg, 1987; Breakey et al., 1989). Homeless mothers (i.e. homeless women with dependent children), however, are seldom married, and have poor work histories (Bassuk et al.,
1986; Smith & North, 1994). Often there is a background of family
violence and of abusive relationships with partners. Smith & North's
(1994) study of 300 homeless women from shelters in St Louis found
that 90% were mothers and that most were dependent on welfare.
In British studies the majority of homeless women have been white
(Drake et al., 1982), although more recently this profile has been
changing. In James' (1991) study of 43 homeless women, 15 (35%)
were black. In another study of 2308 pregnant women who booked
into an antenatal clinic at a London teaching-hospital, over a 1 year
period, 185 (8%) were found to be homeless (Paterson & Roderick,
1990). Compared to a housed group, the homeless group included a
higher proportion of Indo-Pakistani women living in bed and
Homeless women
67
68
E. J. MARSHALL
Men: Women
Men
Women
Adams (1991)
Scott (1991)
James (1991)
Marshall & Reed (1992)
0:64
0:46
0:43
0:70
42%
19%
21%
64%
(27)
(9)
(9)
(45)
Hospital studies
Fernandez (1984)
Herzberg (1984)
115:28
62:48
33.3%
25.8%
34.7%
41.7%
12.1%
42.0% a
85%
17.1%
48.7% a
1.0-12.5%b
American studies
65:13
125:78
0:300
Homeless women
69
70
E. J. MARSHALL
Study
a
Adams (1991)
James (1991)b
Marshall & Reed
(1991)c
Herzberg (1984)
Fernandez (1984)
Breakey^a/. (1989)a
Smith & North (1994)
Men
(%)
0:64
0:43
0:70
62:48
115:28
125:78
0:300
Women (%)
8
7
36
35.5
46.0
56
8.4
4.1
16.5
Non-mothers: 19.4%
Children present: 12.7%
Children not present: 33.3%
Children over 16: 12.5%
Alcohol dependence.
Alcohol seen as problem by staff.
c
Drinking heavily.
Homeless women
71
patterns of drug and alcohol misuse have changed since, thus the
findings cannot be generalized to the present.
In Herzberg's (1987) retrospective study of no fixed abode admissions to a psychiatric hospital, 8.4% of women and 35.5% of men
received a diagnosis of alcoholism. Fernandez (1984) reported similar
figures, 4.1% for women and 46% for men. Breakey et al.'s (1989)
Baltimore sample was assessed almost a decade later. Initially the
Short Michigan Alcohol Screening Test (SMAST: Selzer et al., 1975)
was used to assess 298 homeless men and 230 homeless women in the
missions, shelters and jails of Baltimore and 69% of men and 38% of
women were found to be definite or probable alcoholics. When a
subsample of 125 men and 78 women was interviewed, 56% of the
men and 16.5% of the women fulfilled DSM-III criteria for the
alcohol dependence syndrome (American Psychiatric Association,
1980).
Marshall & Reed (1992) reported that 25 (36%) of 70 women
drank heavily and that 7 (10%) admitted to a current or serious drug
problem. Seven (10%) had a history of opiate dependence and three
were injecting at the time of the interview.
72
E. J. MARSHALL
mental morbidity was twice that for the region as a whole. The
homeless population had higher utilization of general practitioner,
accident and emergency departments and inpatient services.
Herzberg (1987) found no difference in the proportion of homeless
men and women with physical illness in his sample of no fixed abode
admissions, and in Fernandez' (1984) sample of a similar group 1.5%
of the men and none of the women were reported as having 'mainly
medical problems'.
Adams (1991) reported that 8 (12%) of 64 homeless women in a
London hostel had some sort of physical illness, including asthma, hip
pain, epilepsy and deafness. In Marshall & Reed's (1992) older
sample, 19 (27%) of 70 women admitted to physical illness (epilepsy,
4; respiratory, 4; alcohol-related, 2; cardiovascular, 2; diabetes, 1;
other, 6).
In the Baltimore study, Breakey et al. (1989) organized a comprehensive physical assessment of 120 homeless men and 75 homeless
women, using a standardized protocol. This included a review of the
past medical history, a physical examination and laboratory tests (an
electrocardiogram, chest X-ray, blood count and liver function tests,
urinalysis, tests for gonorrhoea and syphilis, tuberculin skin test and
stool examination for parasites). Men had on average 8.3 problems
and women 9.2, and each problem was considered sufficient for
referral to primary care. Problems of the mouth and teeth (orodental)
were the commonest in both sexes. Two thirds of women had
gynaecological problems and some kind of arthritis was diagnosed in
32% of women and 26% of men. Anaemia was found in 35% of
women and 18% of men.
Homeless women
73
74
E. J. MARSHALL
Homeless women
75
Conclusions
Homeless women are a heterogenous group, largely younger and
more socially stable than homeless men. Proportionally more have
major mental illness than men. Older homeless women have high
levels of schizophrenia while personality disorder and substance
misuse are commoner in younger women. They are restless rather
than rootless and can engage with appropriate services. Although the
numbers of homeless women are rising, information is still sparse and
more research on their problems and needs is required.
References
Adams, C. E. (1991). Homeless Women: a Prevalence Study of
Homeless Women in London. Unpublished MSc. thesis. London
School of Hygiene and Tropical Medicine.
American Psychiatric Association (1980). Diagnostic and Statistical Manual
of Mental Disorders, 3rd edit. Washington DC: American Psychiatric
Association.
Arce, A. A., Tadlock, M., Vergare, M. J. & Shapiro, S. H. (1983). A
psychiatric profile of street people admitted to an emergency
shelter. Hospital and Community Psychiatry, 34, 812-17.
Bachrach, L. L. (1984). Deinstitutionalization and women. American
Psychologist, 10, 1171-7.
Bassuk, E. L., Rubin, L. & Lauriat, A. (1984). Is homelessness a mental
health problem? American Journal of Psychiatry, 141, 1546-9.
Bassuk, E. L., Rubin, L. & Lauriat, A. (1986). Characteristics of
sheltered homeless families. American Journal of Public Health, 76,
1097-101.
Blumberg, L. U., Shipley, T. E. & Barsky, S. F. (1978). Liquor and
Poverty: Skid Row as a Human Condition. New Brunswick: Rutgers
Centre of Alcohol Studies.
Brandon, D. (1973). Community for homeless women. Social Work
Today, 4, 167-70.
Breakey, W. R., Fischer, P. J., Kramer, M. et al. (1989). Health and
mental health problems of homeless men and women in Baltimore.
Journal of the American Medical Association, 262, 13527.
Burt, M. R. & Cohen, B. E. (1989). America's Homeless: Numbers,
Characteristics and Programs that Serve Them. Washington DC: Urban
Institute.
Central London Outreach Team (1984). Sleeping out in Central London.
London: Greater London Council.
Chesterton, C. (1926). In Darkest London. London: Stanley Paul and Co.
Ltd.
76
E. J. MARSHALL
Digby, P. W. (1976). Hostels and Lodging Houses for Single People. London:
160-5.
Grella, C. (1994). Contrasting a shelter and day centre for homeless
mentally ill women: four patterns of service use. Community Mental
Health Journal, 30, 3-16.
Kroll, J., Carey, K., Hagedorn, D., Dog, P. F. & Benavides, E. (1986).
A survey of homeless adults in urban emergency shelters. Hospital
and Community Psychiatry, 37, 283-6.
Laidlaw, S. I. A. (1956). Glasgow Common Lodging-Houses and the People
Homeless women
11
6
Homelessness and criminality
PHILIP JOSEPH
Introduction
The relationship between homelessness, mental disorder and criminal behaviour is complex. Although many homeless people are not
mentally disordered and have no criminal history, there is a strong
association between homelessness and mental disorder on the one
hand and homelessness and criminality on the other. The link
between mental disorder and crime is more tenuous and is outside the
scope of this chapter although a comprehensive review is provided by
Wessely & Taylor (1991).
Theories of causation examining homelessness, mental disorder
and crime vary according to ideological standpoint. Whilst criminological theories stress social and economic factors contributing to
homelessness such as unemployment, lack of low cost housing and
poverty, which in turn lead to mental disorder and crime, a
psychological approach tends to focus on individual, intra-psychic
factors, for example the onset of mental illness that precipitates
homelessness and thence crime. Regardless of ideological perspective,
homelessness provides the link between mental disorder and crime,
and once an individual has become homeless, for whatever reason,
the chances of being overlooked by mental health services or
becoming ensnared in the criminal justice system are greatly increased.
This chapter will consider the plight of that most disadvantaged
and chaotic group, namely those homeless people who are both
mentally disordered and criminal. The historical review shows how
homelessness has been criminalized since the Middle Ages as a means
of social control. The review of studies of homelessness and criminal78
79
ity supports the strong links that exist with high rates of homelessness
amongst prison and psychiatric hospital populations. Contemporary
problems brought about by changes in mental health care policy over
the last 30 years will be considered, particularly the increasing role of
the policy as the main community response to disturbed behaviour.
Finally some solutions will be offered, which in the light of the UK
Reed Report (Department of Health and Home Office, 1992) will
stress the early diversion of the mentally disordered from the criminal
justice system as the first part of a package of measures in the
community, such as access to primary care and a possible community
treatment order, as a way of intervening in the cycle of disadvantage.
Historical background
The interaction between the law, social policy and public sentiment is
clearly evident in society's response to the single homeless person. The
fundamental quandary is whether the vagrant should be treated as a
pauper or a criminal. Homelessness itself has been criminalized to
varying degrees in the UK since the Middle Ages. Relief for the poor
of the parish has always taken precedence over relief for poor people
coming from outside the community. The poor, whether sick, frail,
old or mentally ill relied on a dwindling proportion of the church
tithes for their relief or on alms given indiscriminately by the wealthy.
To leave the parish was considered a grave threat to society and
attracted punitive legislation. The early statutes aimed to suppress
vagrancy by punishment without any attempt at relieving the
destitution that gave rise to it.
In the wake of the Napoleonic Wars, a combination of returning
soldiers, the profound social changes set in motion by the beginnings
of industrialization and the rapidly rising population, led to a
dramatic increase in vagrancy. Workhouses, which had been set up
sporadically during the eighteenth century were overcrowded and
disease ridden. In 1821 it was estimated there were at least 60000
people perpetually circulating up and down the country (RibtonTurner, 1887, p. 232). The policy of returning them to their parish of
origin under the Law and Settlement and Removal had become an
expensive failure, open to abuse.
The Vagrancy Act 1824 entitled 'An Act for the Punishment of idle
and disorderly Persons, and Rogues and Vagabonds...' abandoned
80
P.JOSEPH
81
the total for England and Wales. The number prosecuted for sleeping
out increased less dramatically, and tended to average one tenth of
the begging prosecutions. During the 1920s and 30s, numbers were
high and fairly constant, but never reached pre-war levels. A graphic
account of homelessness between the wars has been provided by
Orwell (1933; and see Chapter 2).
After the Second World War, prosecutions dropped markedly and
the numbers of homeless sleeping out almost disappeared, partly due
to the National Assistance Act of 1948. This led the then London
County Council to suspend its annual homeless census. In the 1980s
there has been a reversal of this trend, and although prosecutions for
sleeping out have remained very low, probably due to a deliberate
policy of not arresting those sleeping rough, the number of prosecutions for begging has begun to rise.
Of perhaps more importance is the propensity of the homeless
single person to commit crimes outside the vagrancy statutes. Such
crimes may themselves lead to the breaking of ties with society and the
descent into an unsettled way of life, or they may be encouraged by
this life-style and in either case may hinder escape from it. In earlier
times the bands of 'sturdy rogues' roaming the country, causing
havoc and spreading panic, were a real menace to society, but today's
single homeless person is likely to be a recidivist petty offender.
82
P.JOSEPH
The mentally ill were next with a rate of 55%, followed by those with
personality disorder (69%), and finally the group with the highest
rate of recorded criminality were the alcoholics at 76%. Tidmarsh
also found that the more handicaps suffered by an individual, e.g.
mental illness complicated by alcohol abuse, the higher the rate of
criminality. The most common offences were theft, criminal damage
and vagrancy; violence was much less likely, although higher than
would be expected for the general population.
Tidmarsh then looked at the hospital careers of his homeless men.
He showed that for those with a diagnosis of mental illness or
alcoholism, a history of imprisonment was associated with previous
psychiatric hospitalization. This finding is supported by a study of
529 homeless adults in Los Angeles, where those with a history of
hospital treatment had higher rates of imprisonment and longer
periods of homelessness than the non-hospitalized homeless (Gelberg
etal., 1988).
Further evidence that confirms the close relationship between
homelessness, mental disorder and criminality is provided by those
studies of prison populations that show high rates of homelessness
amongst prisoners prior to their incarceration, and studies of homelessness amongst psychiatric hospital admissions. Turning to the
prison studies first; an early study of prisoners in the south-west of
England found that of 174 sentenced men, 31 % were homeless at the
time of their arrest, and 42% expected to have no home to go to on
their release. Those with no homes were older, more recidivist with
higher rates of unemployment than the rest (South-West Regional
Group Consultative Committee, 1969). A more extensive study
covering London and south-east England showed higher rates of
homelessness (Walmsley, 1972). He estimated that 7000 (58%) of the
12000 prison releases annually were of men who were possibly
homeless. They had either been sleeping rough or in common lodging
houses for part of the year prior to their arrest. The homeless prisoners
were older and serving shorter sentences, they were more likely to be
single or separated and originate from Scotland or Ireland. In
another survey of prisoners, Gibbens & Silverman (1970) showed
that those who were alcoholic had three times the rate of homelessness
than the other prisoners.
Turning to hospital admissions, Berry & Orwin (1966) retrospectively surveyed the admission of homeless patients to a Birmingham
psychiatric hospital between 1961 and 1965. The percentage of male
83
84
P.JOSEPH
Current problems
In England and Wales, a policy of community care has existed for
over 30 years, during which there has been a decline in the mental
hospital and mental handicap hospital population and a rise in the
remand and sentenced prisoner population. From 1962 to 1990, the
combined psychiatric inpatient population fell from 196234 to
87 683, a 55% reduction; the combined prison population rose from
31063 to 47 936, an increase of 54% (Home Office, 1962-1990;
DHSS 19621990). The percentage changes are similar but the gross
numbers differ widely, thus invalidating some of the wilder claims
made about the wholesale transfer of the mentally ill from hospital to
prison (Weller, 1989). However, when one narrows the focus on to the
disadvantaged world of the homeless mentally ill, there is perhaps
justified concern that many, who previously would have been
admitted to psychiatric hospitals, are now becoming ensnared in the
criminal justice system and ending up in prison.
The police are becoming increasingly recognized as an important
community mental health resource, particularly as the focus of
treatment for the mentally ill moves to the community. They are the
only laymen in Great Britain who can act alone, without medical
evidence, to detain a suspected mentally ill person in a place of safety.
This power has existed in statutory form for over 100 years and the
latest version is contained in section 136 of the Mental Health Act
1983, which states:
1. If a constable finds in a place to which the public have access a
person who appears to him to be suffering from mental disorder
and to be in need of care and control, the constable may, if he
thinks it necessary to do so in the interests of that person or for the
safety or protection of other persons, remove that person to a place
of safety...
2. A person removed to a place of safety under this section may be
detained there for a period not exceeding 72 hours for the purpose
of enabling him to be examined by an Approved Social Worker
and of making any necessary arrangements for his treatment and
care.
This power coupled with the police's round-the-clock availability to
intervene in emergencies, makes it hardly surprising that they
85
become involved with the mentally ill, whether they see this as part of
their job or not.
There is great variation across the country in the number of police
referrals to psychiatric services. High rates are especially common in
inner city areas and police in London are particularly likely to use
their powers under section 136 Mental Health Act 1983, rather than
rely on the attendance of approved social workers or other mental
health professionals at the police station. In 1986, for example, 47%
of all section 136 admissions in England took place in London. Rollin
(1965) has attributed this excess in London to the city's role as a
'sump ... into which chronic psychotics from all over the United
Kingdom ... are drained'.
The reasons that led the mentally ill to come to the attention of the
police were identified by George (1972) in his study of police referrals
to psychiatric services in west central London. Threatening or bizarre
behaviour leading to public disturbance predominated, followed by
suicide attempts or threats. George showed that those dealt with
under section 136 differed from those admitted compulsorily under
emergency civil provisions, i.e. section 29 of the then Mental Health
Act 1959, which is now covered by section 4 of the 1983 Act. The
police admissions had significantly more often shown aggressive
behaviour before admission and presented more management problems during admission. They were more likely to be of no fixed abode
and socially disorganized. Interestingly, those admitted compulsorily
in the past tended to have been on the same section as the index
admission.
Numerous studies have examined the characteristics of police
referred psychiatric admissions in the UK and the USA and certain
conclusions can be drawn. Police referrals have high rates of
chronic serious mental illness, frequently requiring emergency compulsory psychiatric admission (Kelleher & Copeland, 1972). Diagnostically, the majority suffer from a psychotic illness, predominantly schizophrenia. For example, in George's (1972) sample of
856 section 136 admissions collected between 1967 and 1969, 57%
had a diagnosis of schizophrenia. Very few referrals are considered
to have no mental disorder, indeed Rollin (1965), suggested all 75
cases in his study had a psychiatric diagnosis. Other studies suggest
a figure closer to 10% as having no mental disorder (Rogers &
Faulkner, 1987; Fahy et al., 1987). In the USA, Sheridan & Teplin
(1981), have confirmed these diagnostic findings in their study of
86
P.JOSEPH
Some solutions
Diversion from custody
When criminal proceedings are brought, the defendant who is
mentally disordered has the same right as other defendants to be
released on bail before trial unless one or more of the exceptions
specified in the Bail Act 1976 apply. In practice the presumption in
favour of bail is weakened for the mentally disordered defendant due
to the clause allowing remand in custody for his own protection. This
is compounded if the defendant is of no fixed abode, as 'lack of
community ties' is cited as a reason for forming the belief that the
defendant would fail to return to court. Thus, mentally disordered
homeless defendants charged with minor offences are more likely to
be remanded into custody than others charged with similar offences.
There has been mounting concern over the conditions endured by
mentally disordered prisoners on remand (Herridge, 1989; HM Chief
Inspector of Prisons, 1990). Suicide in prison is a matter of grave
concern; between 1972 and 1989 there were 295 suicides in prisons in
England and Wales with an increase in the yearly rate far in excess of
the rise in the prison population (Dooley, 1990). In order to try and
reduce the numbers of mentally ill prisoners the Home Office has
issued guidelines to all courts regarding provision for mentally
disordered defendants, which includes the statement: 'A mentally
disordered person should never be remanded to prison simply to
receive medical treatment or assessment' (Home Office, 1990).
It is not uncommon for there to be an inordinate delay following
87
88
P.JOSEPH
89
90
P.JOSEPH
91
Conclusion
When homelessness is placed in its historical context it can be seen
that its relationship with crime has always been close with the fate of
the homeless person entwined with that of the criminal. For many
92
P.JOSEPH
93
lurching from one extreme to the other, whilst the continual victim,
the homeless, mentally disordered offender, is caught in a locked
revolving door.
References
Allderidge, P. (1979). Hospitals, madhouses and asylums: cycles in the
care of the insane. British Journal of Psychiatry, 134, 321-4.
Berry, C. & Orwin, A. (1966). No fixed abode: a survey of mental
hospital admissions. British Journal of Psychiatry, 112, 1019-25.
Bowden, P. (1978a). Men remanded into custody for medical reports. I.
The selection for treatment. British Journal of Psychiatry, 132, 32031.
Bowden, P. (1978b). Men remanded into custody for medical reports.
II. The outcome of the treatment recommendation. British Journal
of Psychiatry, 132, 332-8.
Burns, T., Goddard, K. & Bale, R. (1993). Mental health professionals
favour community supervision orders. British Medical Journal, 307,
803.
Coid, J. (1988a). Mentally abnormal prisoners on remand. I. Rejected
or accepted by the NHS. British Medical Journal, 296, 1779-82.
Coid, J. (1988b). Mentally abnormal prisoners on remand. II.
Comparison of services provided by Oxford and Wessex regions.
British Medical Journal, 296, 1783-4.
Dell, S., Grounds, A., James, K. & Robertson, G. (1991). Mentally
Disordered Remanded prisoners. Unpublished report to the Home
Office.
Department of Health and Home Office (1992). Review of Health and
Social Services for Mentally Disordered Offenders and Others Requiring
Similar Services. (Reed Report). London: HMSO. (CM 2088).
DHSS (Department of Health and Social Security) (1962-1990). Health
and Personal Social Services Statistics for England (With Summary Tables
for Great Britain). Annual Reports. London: HMSO.
Dooley, E. (1990). Prison suicide in England and Wales, 1972-87.
British Journal of Psychiatry, 156, 40-5.
Eastman, N. (1994). Mental health law: civil liberties and the principle
of reciprocity. British Medical Journal, 308, 43-5.
Edwards, G., Williamson, V., Hawker, A., Hensman, C. & Postoyan, S.
(1968). Census of a reception centre. British Journal of Psychiatry,
114, 1031-9.
El-Kabir, D. (1982). Great Chapel Street medical centre. British Medical
Journal, 284, 480-1.
Fahy, T., Bermingham, D. & Dunn, J. (1987). Police admissions to
psychiatric hospitals: a challenge to community psychiatry.
Medicine, Science and the Law, 27, 263-8.
Gelberg, L., Linn, L. & Leake, B. (1988). Mental health alcohol and
94
P.JOSEPH
95
Rollin, H. (1969). The Mentally Abnormal Offender and the Law. Oxford:
Pergamon Press.
Rollin, H. (1993). Community care: Italy's 'U' turn. Psychiatric Bulletin,
17, 494-5.
Royal College of Psychiatrists (1993). Community Supervision Orders.
London: RCP.
Royal Commission (1993). The Royal Commission on Criminal Justice Report.
London: HMSO. (Cm 2263).
Sensky, T., Hughes, T. & Hirsch, S. (1991a). Compulsory psychiatric
treatment in the community. I. A controlled study of compulsory
community treatment with extended leave under the Mental
Health Act: special characteristics of patients treated and impact of
treatment. British Journal of Psychiatry, 158, 792-98.
Sensky, T., Hughes, T. & Hirsch, S. (1991b). Compulsory psychiatric
treatment in the community. II. A controlled study of patients
whom psychiatrists would recommend for compulsory treatment in
the community. British Journal of Psychiatry, 158, 799-804.
Sheridan, E. & Teplin, L. (1981). Police-referred psychiatric
emergencies: advantages of community treatment. Journal of
Community Psychology, 9, 140-7.
Sims, A. & Symonds, R. (1975). Psychiatric referrals from the police.
British Journal of Psychiatry, 127, 171-8.
South-West Regional Group Consultative Committee (1969). The
Homeless Offender in the South-west of England. South-West
Regional Group Consultative Committee for After-Care Hostels.
Szmukler, G., Bird, A. & Button, E. (1981). Compulsory admissions in a
London borough. I. Social and clinical features and a follow-up.
Psychological Medicine, 11, 617-36.
Tidmarsh, D. (1977). Psychiatric Disorder in a Population of Homeless
Destitute Men. MD thesis. University of Cambridge.
Walmsley, R. (1972). Steps from Prison. London: Inner London
Probation and After-Care Services.
Weller, M. (1989). Mental illness - who cares? Nature, 339, 249-52.
Wessely, S. & Taylor, P. (1991). Madness and crime: criminology versus
psychiatry. Criminal Behaviour and Mental Health, 1, 193-28.
PART II
SERVICES
7
Services for the mentally ill homeless
E. JANE MARSHALL AND DINESH BHUGRA
Introduction
As for estimates of homeless populations in various countries (see
Chapters 10, 12, 13 and 14) the estimates of various types of mental
illness and other psychological morbidity in homeless populations
vary. Most of the research so far has been in the 'captive' samples that
are easy to approach and identify and findings from these populations
cannot be generalized. We know from the existing literature that
most homeless individuals do not receive help from any service
agency and their distrust of statutory health services and the
inflexibility of these services in return are important factors in
continuing poor physical and mental conditions.
Between one third and one quarter of the homeless population
suffers from serious mental illness (Tessler & Dennis, 1989; Dennis et
al., 1991; see Bhugra 1993, for a review). The relationship between
homelessness and mental illness is complex and mental illness is only
one of many interacting factors leading to and perpetuating homelessness. In addition, other factors include the reduction in availability of low cost housing, increasing unemployment, loss of the
manufacturing base (in the UK) and closure of direct access hostels,
which traditionally served as asylums for large numbers of the
mentally ill (Craig & Timms, 1992; see Chapters 2 and 3). Cohen &
Thompson (1992) argue that the distinction between the homeless
mentally ill and the mentally ill homeless is not a semantic one but is
vital in terms of addressing needs and in the development of services.
They challenge the assumption that the fundamental problem of the
homeless mentally ill is their mental illness and emphasize that the
conceptual dichotomy between homeless persons with or without
99
100
mental illness is purely illusory. 'Not mentally ill' homeless people are
likely to have many mental health problems and the 'mentally ill'
homeless may have non-psychiatric problems that arise from the
socio-political elements affecting all homeless people. Their model
views homelessness as the core element of the broader socio-economic
and political context that in turn becomes intertwined with personal
biography and illness.
Deinstitutionalization is often blamed for the increase in homelessness in the USA, but research data on this issue remain sparse (Fischer
& Breakey, 1986; Dennisetal., 1991; Cohen & Thompson, 1992). The
British experience suggests that adequate follow-up and support for
long-stay psychiatric patients discharged from hospital is critical in
preventing homelessness (Dayson, 1992).
Most studies on the homeless mentally ill have been, until relatively
recently, cross-sectional and lacking in methodological rigour. Longitudinal studies are needed to assess the extent to which individuals
move in and out of homelessness and to evaluate the contributory
impact of mental illness.
101
Services
The homeless mentally ill do not access psychiatric and substance
abuse services, despite needing them (Padgett etal., 1990). Breakey et
al. (1989) reported that only 47% of 298 homeless men and 30% of
230 homeless women were able to name a usual source of health care.
However, 20% overall had been hospitalized for a physical illness in
the previous year and 23% of men and 33% of women had been
psychiatric inpatients over the same period. Victor (1992) reported
that 92% of 319 temporarily homeless people resident in bed and
breakfast hotels in North West London were registered with a general
practitioner. This sample was predominantly female (61%) and
young (72% between the ages of 16 and 34 years). One tenth had had
an acute health problem in the previous fortnight, 46% had been
experiencing a health problem for 1 year or more and 34% had a
health problem that had been limiting their daily activity for 1 year or
more.
Factors related to service utilization amongst the homeless are
poorly understood and little research has been carried out in this area.
Higher rates of hospitalization have been reported among homeless
women (Robertson, 1986; Grella, 1994; see Chapter 5). This may be
due to the fact that homeless women are more likely to be admitted to
hospital whereas homeless men in the same situation may be taken to
prison (Herzberg, 1987).
102
Specific problems
Models of mental health care delivery
The emphasis on catchment areas and sectorization within catchment areas not only excludes individuals with no address but also
those who are not interested in long-term follow-up. The geographical mobility of the homeless does not recognize catchment
area boundaries and follow-up between catchment areas is often
lacking.
Attitudes
Mental health professionals find it difficult to deal with the multiple
complex problems of the homeless mentally ill. Many have been
reluctant to work with the homeless, and the perceived dichotomy of
the 'deserving' and the 'undeserving' poor has led to negative
attitudes. Bachrach (1990) observed that the attitudes of the media
towards the homeless do affect the services. In places where libraries
and railway stations are used by the homeless, joint local planning has
sometimes helped to improve services (Petroskey, 1988). Innovative
and joint working between mental health workers and other organiz-
103
104
The vacuum left by the failure of mental health services to cater for
anything other than the narrowly defined psychiatric needs of the
homeless mentally ill has been filled by voluntary groups who provide
shelter, food and support. However, the staff in these agencies,
although qualified in other ways, may have had little training in
mental health assessment or management. Lack of resources (Fischer
&Breakey, 1986) and overwhelming demands (Marshall, 1989) have
often put voluntary agencies under a tremendous strain. In addition,
minimal liaison between various voluntary agencies and between
voluntary and statutory groups has meant that a comprehensive
mental health service for the homeless mentally ill cannot be
delivered.
What is required?
Community mental health programmes must be adapted to meet the
range of needs of the mentally ill homeless. The degree of disability of
individuals, their personal strengths, vulnerabilities, preferences and
105
106
services
107
Conclusion
The provision of comprehensive services for the mentally ill homeless
is time-consuming and complex. However, it is also an opportunity to
be innovative and flexible and may, in time, serve to inform the
development of the standard mental health services.
References
Bachrach, L. L. (1987). Geographical mobility and the homeless
mentally ill. Hospital and Community Psychiatry, 38, 2 7 - 8 .
108
109
8
Clinical work with homeless people in the
USA
WILLIAM R. BREAKEY
111
112
W. R. BREAKEY
treatment, even when they are unable to consent to it, and that
admitting a person to a substandard facility is the lesser of two evils,
when compared to having that person stay on the streets (Lamb et al.,
1992b). However, the majority of those who work in the field are of
the opinion that the great majority of mentally ill people can be
treated or sheltered without resorting to involuntary treatment
procedures, and that, instead of looking to institutional models of
care, policy makers' attention should be focused upon the great
poverty of many mentally ill people, the lack of suitable housing and
the need for better community support systems, which might prevent
their homelessness. What is more, Appelbaum (1992) points out that
the literature on involuntary hospitalization provides little evidence
that creating more draconian commitment laws would have the effect
of reducing homelessness among the mentally ill.
The consumer movement has also had considerable impact on
thinking about services for homeless people in recent years, as it has in
other areas of public life and in community mental health. Organizations of homeless and formerly homeless people have had important
roles, through advocacy and confrontation, in sensitizing the public
and those in power to the dilemmas homeless people face. Consumerrun programmes have been able to reach out to homeless people in
ways that may be more effective than conventional professional
approaches (Van Tosh, 1993).
113
114
W. R. BREAKEY
115
o/DSM-III
Women
9.0
7.3
16.3
10.3
4.7
1.3
9.7
9.3
12.6
6.3
13.8
2.3
8.3
2.0
8.3
1.8
67.0
28.8
26.2
24.7
30.0
46.5
24.3
46.2
45.3
11.1
SMI, severely mentally ill: mental illness diagnosis, with history of extensive
hospitalization and/or severe functional impairment.
a
Includes paranoid disorder, atypical psychosis, organic hallucinosis, etc.
116
W. R. BREAKEY
ence, which was found in one quarter of the Baltimore sample, creates
major problems for treatment and rehabilitation.
These data were obtained from men and women surveyed in the
general shelter system and the city jail in Baltimore. Other subgroups
of homeless people may have different profiles of disorders. For
example, a survey of mothers in shelters for families in Massachusetts
revealed a different profile in these young women. Personality
disorders were more frequently diagnosed, major mental illnesses less
frequently (Bassuk et al., 1986). Several studies of women have found
high prevalence rates of post-trauma tic stress disorder (Smith et al.,
1993). Prevalence rates may also vary over time. A recent survey of
homeless women in Baltimore has shown a massive increase in
cocaine abuse between the mid-1980s and the early 1990s, in keeping
with trends in the general population.
117
118
W. R. BREAKEY
119
120
W. R. BREAKEY
Clinical considerations
What are the practical implications for the clinician? What special
understanding or techniques are needed to work with homeless
people? As the words 'homeless mentally ill' indicate, there are two
broad strategic objectives: to treat the illness and to end the
homelessness. At each stage, both objectives need to be kept in mind.
Without treating the illness, the person is unlikely to be able to obtain
or retain housing; without addressing the housing problem, attempts
at treatment of the mental illness will be futile.
Often a clinician must exercise great patience to establish trust.
Simply offering a friendly word, or some food or clothing, may be all
that is possible at first. Over time, as trust develops, more active
interventions become possible. Susser (1992) took many months to
gain the confidence of both staff and residents of the residence in
which he was trying to provide service. The role he identified as one
that would permit him to move into the hotel and be accepted
initially was that of organizer of a weekly Bingo game. This role was
sufficiently unthreatening to the women in the hotel for him to
gradually gain their confidence enough to move into other areas.
Social workers with Project Outreach at the Goddard Riverside
Community Center in New York were the first to adopt a trustbuilding technique that has since been adopted widely. Each day
they would make trips out into their community with a store of
sandwiches to distribute and to exchange a greeting or engage in a
brief conversation with the men and women they encountered. They
took months on occasion to establish contact with some of the most
timid of the street people; many contacts were often needed before the
individual gained sufficient confidence in the workers to establish
some sort of tenuous relationship.
In many cases, homeless patients are among the most severely and
chronically ill that a therapist is likely to encounter. They are
'treatment resistant' and have minimal resources in material terms as
well as in terms of their social environment. To avoid disillusionment,
it is wise to keep in mind that the prognosis in many cases is poor.
Some homeless people may be reluctant to contemplate change in
121
122
W. R. BREAKEY
accessibility
123
124
W. R. BREAKEY
125
126
W. R. BREAKEY
The future
Those who engage in providing services for homeless people do so in
the hope that their services will soon no longer be needed, because
homelessness will cease to be a major problem in American society.
However this hope seems increasingly naive. As the years pass, the
numbers of homeless people seeking help from service agencies only
increases. The enormous deficiency in the supply of housing in
America may take decades to remedy. Public opinion is not supportive of significant increases in assistance to the poor. Education for the
masses of young Americans is not fitting them well for employment in
an increasingly technological employment market.
America is struggling to reform its health care system. In spite of
127
128
W. R. BREAKEY
129
References
Appleby, L. & Desai, P. (1987). Residential stability: a perspective on
system imbalance. American Journal of Orthopsychiatry, 57, 515-24.
Appelbaum, P. S. (1992). Legal aspects of clinical care for severely
mentally ill, homeless persons. Bulletin of the American Academy of
Psychiatry and the Law, 20, 455-73.
Ball, F. L. J. & Havassy, B. E. (1984). A survey of the problems and
needs of homeless consumers of acute psychiatric services. Hospital
and Community Psychiatry, 35, 97-9.
Bassuk, E. L., Rubin, L. & Lauriat, A. S. (1986). Characteristics of
homeless sheltered families. American Journal of Public Health, 76,
1097-101.
Bebout, R. R., Harris, M., Swayze, F. V. et al. (1993). The Community
Connections Social Support Network Intervention Model. Washington DC:
Community Connections, Inc.
Billig, N. & Levinson, C. (1987). Homelessness and case management in
Montgomery County, Maryland: a focus on chronic mental illness.
Psychosocial Rehabilitation Journal, 11, 59-66.
Blackwell, B., Breakey, W. R., Hammersley, D., Hammond, R.,
McMurray-Avila, M. & Seagar, C. (1990). Psychiatric and mental
health services. In P. W. Brickner, L. K. Sharer, B. A. Conanan,
M. Savarese & B. Scanlan (eds.), Under the Safety Net: the Health and
Social Welfare of the Homeless in the United States, pp. 184-203. New
York: Norton.
Breakey, W. R. & Fischer, P. J. (1995). Mental illness and the
continuum of residential stability, Social Psychiatry and Psychiatric
Epidemiology, 30, 147-51.
Breakey, W. R., Fischer, P. J., Kramer, M. et al. (1989). Health and
mental health problems of homeless men and women in Baltimore.
Journal of the American Medical Association, 262, 1352-7.
Brickner, P. W., Sharer, L. K., Conanan, B. A., Savarese, M. &
Scanlan, B. (eds.) (1990). Under the Safety Net: The Health and Social
Welfare of the Homeless in the United States. New York: Norton.
Burt, M. R. & Cohen, B. E. (1989). America's Homeless. Washington,
DC: The Urban Institute.
Center for Mental Health Services (1994). Making a Difference: Interim
Status Report of the McKinney Demonstration Program for Homeless Adults
with Serious Mental Illness. Washington, DC: US Department of
Health and Human Services.
Cohen, C. I. (1993). Poverty and the course of schizophrenia:
implications for research and policy. Hospital and Community
Psychiatry, 44, 951-8.
Cohen, N. L. & Marcos, L. R. (1986). Psychiatric care of the homeless
mentally ill. Psychiatric Annals, 16, 729-32.
Dixon, L., Kraus, N. & Lehman, A. (1994). Consumers as providers:
the promise and the challenge, Community Mental Health Journal, 30,
615-25.
130
W. R. BREAKEY
1041-5.
Farr, R. K. (1986). A mental health treatment program for the
homeless mentally ill in the Los Angeles Skid Row area. In B. E.
Jones (ed.), Treating the Homeless: Urban Psychiatry's Challenge,
1973-9.
Goetcheus, J., Gleason, M. A., Sarson, D., Bennett, T. & Wolfe, P. B.
(1990). Convalescence: for those without a home - developing
respite services in protected environments. In P. W. Brickner, L. K.
Sharer, B. A. Conanan, M. Savarese & B. Scanlan (eds.), Under the
Safety Net: the Health and Social Welfare of the Homeless in the United
131
132
W. R. BREAKEY
Introduction
The terms 'shelter', 'hostel' and 'lodging house' have been used
interchangeably by most UK commentators on homelessness. Unfortunately the imprecise use of these terms has obscured important
differences between two distinct types of accommodation that have
traditionally been provided for the single homeless in the UK. In this
chapter these two types of accommodation will be referred to as
shelter accommodation and hostel accommodation. If we are to make
sense of the UK studies of homeless people with mental disorder it is
necessary to reassert the distinction between shelter and hostel
accommodation. Consequently this chapter will begin by defining
hostel and shelter accommodation. Studies of homeless people in the
UK will then be classified into hostel or shelter populations. The
findings revealed by this reclassification will then be discussed.
Finally, on the basis of these findings, the size of the mentally ill hostel
population, and the characteristics of its members, will be discussed.
The effectiveness of hostels for the homeless in caring for people with
severe psychiatric disorder is discussed in Chapter 16.
accommodation
134
M. MARSHALL
135
136
M. MARSHALL
City
Date
Laidlaw (1956)
LCSS (1960)
Priest (1971)
Crossley & Denmark (1969)
Lodge Patch (1971)
Tidmarsh & Wood (1972)
Leopoldt & Lynch (1981)
Timms & Fry (1989)
Marshall (M) (1989)
Marshall (J) & Reed (1992)
Reed et al. (1992)
Stark et al. (1989)
Adams et al. (1996)
Geddes et al. (1994a)
Newton et al. (1994)
Weller et al. (1989)
Glasgow
London
Edinburgh
Bolton
London
London
Oxford
London
Oxford
London
London
Various centres
London
Edinburgh
Edinburgh
Various centres
1954
1955
1965
1969
1970
1971
1979
1987
1988
1988
1988
1989
1990
1992
1992
1985
Prevalence of
schizophrenia (%)
3
5
32
24
15
16
17
31
27
64
13
18-25
50
9
3
22a
Note: the date column in the table records the year when the study was begun,
not the year of publication.
a
N o estimate for schizophrenia was given in this study, but 22% had 'active
psychotic symptoms'.
Hostel studies
Eight of the studies listed in Table 9.1 are unequivocally studies of
hostel populations: a study of Glasgow Common Lodging houses
(Laidlaw, 1956); a study of London Common Lodging Houses
(LCSS, 1960); a study of Edinburgh Common Lodging houses
(Priest, 1971); and three recent studies: one of hostels in Oxford
(Marshall, 1989), and the other two of women's hostels in London
(Marshall & Reed, 1992; Adams et aL, 1996). In addition, the study
by Crossley & Denmark (1969) of a Salvation Army hostel, unlike the
study of Lodge Patch (1971), is considered to be a true hostel survey,
because the hostel was 'a full board hostel ... the population being
mainly long-stay'.
Mixed studies
In addition there are three studies that have reported the prevalence
of mental disorder in accommodation that was functioning as a
shelter for some persons and as a hostel for others. The first of these
studies reported the rates of mental disorder amongst casual users and
recognized residents of the Camberwell Reception Centre (Tidmarsh
& Wood, 1972). Recognized residents were said to differ from the
casual users in that residents: had agreed to stay for a period of time;
worked in the Centre (cleaning or preparing food, or working in the
workshops); ate separately from the 'casuals'; and could stay in the
Centre during the day and have a midday meal. The second of these
studies compared the rate of mental disorder amongst new admissions
to a Salvation Army Hostel with that amongst long-term residents
(Timms & Fry, 1989). The third study was a random sample of
residents of nine Edinburgh hostels (Geddes et al., 1994a), in which
approximately 10% of the beds were short-term 'shelter accommodation' (J. Geddes, personal communication). In view of the small
number of shelter beds in this study it will be classified as a hostel
study.
137
138
M. MARSHALL
Study
Laidlaw (1956)
LCSS (1960)
Priest (1971)
Grossley & Denmark
(1969)
Tidmarsh & Wood (1972)
Timms & Fry (1989)
Marshall (M) (1989)
Marshall (J) & Reed
(1992)b
Adams et al. (1966)b
Geddes et al. (1994a)
Lodge Patch (1971)
Tidmarsh & Wood (1972)
Leopoldt & Lynch (1981)
Timms & Fry (1989)
Reed et al. (1992)
Newton et al. (1994)
Date
begun
Sample
size
Percentage with
schizophrenia
800
79
55
3
5
32
24
1971
1987
1988
1988
63 a
58
146
70
29
38
27
64
1990
1992
1970
1971
1979
1987
1988
1992
64
136
123
p
76
65
96
65
50
9
15
16
17
25
13
3
1954
1955
1965
1969
There is some ambiguity about the sample size in the original paper, this figure
may be inaccurate.
b
Study carried out in women-only hostels. In other studies subjects were
predominantly or entirely male.
Unclassified studies
Three of the studies listed in Table 9.1 cannot be classified as studies of
shelter or hostel accommodation. In one study this is because data
collected over several years from shelter and hostel populations have
been combined (Weller et al., 1989). This study is therefore omitted
from the subsequent analysis. In another study, of the users of
reception centres (Stark et al., 1989), subjects were selected on the
grounds of being 'hard to place', therefore the study does not provide
a representative sample of the users of the centres. This study will also
be omitted from the following analysis. The final study (Newton et al.,
1994) is the only UK prevalence study of mental disorder amongst
rough sleepers. This study cannot be classified as a shelter study, but
its findings are of interest and will be shown alongside the shelter and
hostel data for comparative purposes.
139
140
M. MARSHALL
3-5
?
?
1961-70
28.7
15
?
1971-80
29
16-17
?
1981-90
41(30.1)
17.6
?
1991a
9
?
3
at about 17%, whilst in the 1960s, 1970s and 1980s the prevalence in
hostels was about twice this figure. These findings, the exact opposite
of accepted wisdom (Scott, 1993), are further reinforced by the
findings of Newton et al. (1994) who (using the same methodology of
Geddes et al., 1994a) found a prevalence of only 3% schizophrenia in
their survey of Edinburgh rough sleepers.
There are four possible explanations of the higher prevalence of
schizophrenia in hostels:
1. Hostels offer care that is attractive to persons with severe mental
disorder. Since hostels cater for semi-permanent residents they
tend to have higher staff ratios and better facilities than shelters.
Thus hostels are better placed to provide support to people
suffering from schizophrenia. Such support could include: getting
meals, assisting with personal hygiene and domestic chores, and
obtaining benefits and managing money.
2. Admission procedures to hostels select people with severe mental
disorders. There is a tradition of using shelter accommodation as a
contact service from which those suitable for hostel accommodation are selected. When this occurs those selected are often
suffering from severe mental disorder. For example, a programme
of action research in the St Mungo's hostels found that homeless
men coming from the street tended to remain in hostels for only
short periods of time. To overcome this problem the researchers
suggested that a 'contact service' should be set up to make contact
with, and to assess, potential residents, before a place was offered.
This new approach was highly successful in increasing length of
stay, but an unforeseen consequence was that amongst those who
settled 'psychiatric disabilities were particularly prominent'
(Leach, 1979). Many hostels are associated with shelters that act
141
142
M. MARSHALL
143
94
Marshall
(1989)
48.1
2
48
3
44%>3yrs
90
12
42%>lyr
100
46
Geddes ea
(1994a)
50.7
42
20
47
95
Hostel females
Shelter males
Reedea (1992)
52.1
30
4
50%>2yrs
40
52
64
18
42
66%>lyr
29
100
11
Data in italics were derived from a sample of mentally ill residents. Data in normal font were derived from a sample of all
residents.
ea, et al.
145
146
M. MARSHALL
Otherwise the social situations of male and female hostel residents are
fairly similar.
There are insufficient data to reach firm conclusions about the
ethnic composition of modern hostel populations. It appears that
people of Scottish and Irish origin are over-represented. There are
also indications that, in London at least, people of West Indian origin
are over-represented. Little recent data is available concerning the
original social class of hostel users. It is of interest however that
Marshall & Reed (1992) found that the fathers of 32 out of 53 women
in two women's hostels were members of social classes 1, 2 or 3.
There are indications that the health of mentally disordered hostel
residents is poor. Several authors have reported physical problems to
be common amongst residents with mental disorder (Stark et aL,
1989; Marshall & Reed, 1992). High mortality rates amongst
mentally disordered hostel residents were suggested in the Oxford
follow-up study described above (Marshall & Gath, 1992).
Psychiatric characteristics
147
References
Adams, C. E., Duke, P. J., Pantelis, C. & Barnes, T. R. E. (1996).
Homeless women: A prevalence study. British Journal of Psychiatry,
(in press).
Balazs, J. (1993). Health care for the single homeless. In K. Fisher (ed.),
Homelessness, Health Care and Welfare Provision, pp. 51-93. London:
Routledge.
Connolly, J. & Crown, J. (eds.) (1994). Homelessness and III Health. A
report of a working party of the Royal College of Physicians. London:
RCP.
Connelly, J. & Williams, R. (1994). Schizophrenia among residents of
hostels for homeless people. British Medical Journal, 308, 1572
(letter).
Crossley, B. & Denmark, J. C. (1969). Community care - a study of the
psychiatric morbidity of a Salvation Army hostel. British Journal of
Sociology, 20, 443-9.
Geddes, J., Newton, R., Bailey, S., Young, G., Freeman, C. & Priest, R.
(1994a). Comparison of prevalence of schizophrenia among
148
M. MARSHALL
149
University Press.
Timms, P. W. & Fry, A. H. (1989). Homelessness and mental illness.
Health Trends, 21, 7 0 - 1 .
10
Old and homeless in London and New York
City: a cross-national comparison
CARL I. COHEN AND MAUREEN CRANE
Introduction
Persons aged 50 and over are estimated to comprise about one fifth of
the homeless in New York City (NYC) and nearly one third of the
homeless in London (Weeden & Hall, 1985; Kelling, 1991). The
proportion of those aged 50 and over living in substandard singleroom occupancy (SRO) hotels or hostels may be even higher
(Keigher, 1991). However, older homeless have been an unwanted
stepchild of both the fields of homelessness and geriatrics. In contrast
to the scholarly literature on younger homeless that has mushroomed
in recent years, there have been relatively few papers written about
ageing homeless (Cohen, 1996). On a service level, most senior citizen
centres have shown little interest in this group, and public shelters
have been typically avoided by older homeless persons because of fear
of aggression by younger homeless, insensitive staff and accommodations that are not suitable for their physical disabilities (Coalition for
the Homeless, 1984). Finally, governmental policy has failed to
recognize the special needs of this ageing group, and even when
statutory entitlements exist, many older homeless have not obtained
such benefits (Kelling, 1991; Crane, 1993).
Although the absolute numbers of homeless persons in London are
lower than in NYC, the former has the highest number of homeless
persons among Western European cities (Adams, 1986; Toro &
Rojansky, 1990; Schmidt, 1992). This commonality of large-scale
homelessness in both cities creates an opportunity to undertake
cross-national comparisons of ageing homeless. Such comparisons
can provide answers to: (a) the proportionate contribution to the
causes of homelessness of individual pathology and behaviour versus
150
151
Definitions
What is coldy among homeless persons?
For the most part, there has been a consensus in the literature that
studies of 'older' homeless should include persons aged 50 and over
(Gelberg et al., 1990). This is because many of these persons at age 50
look and act like persons 10-20 years older in the general community.
However, there are important differences between the 'younger5
homeless (i.e. ages 50 to 60 or 65) and the 'older' homeless (i.e. ages 65
and over). The latter, especially in the USA, are entitled to
considerably more social and health benefits, and this may have
important implications for their well-being.
What is homelessness?
The definition of homelessness in Britain and the USA has represented a compromise between social justice and social expediency
(Hopper, 1991). Until about 1980, homelessness in the USA included
persons living in marginal housing such as SRO hotels and low-cost
boarding houses (Bogue, 1963; Wallace, 1968). With the dramatic
expansion of homelessness that occurred after 1980, homelessness
became more narrowly defined, sometimes to include only street
persons but usually encompassing those persons living in public or
private shelters (Federal Task Force on Homelessness and Severe
Mental Illness, 1992). In Britain, a broader definition of homelessness
is generally used that comprises persons in hostels, bed and breakfast
hotels, squats and on the streets ('sleeping rough') (Gay, 1989; Gay &
Greener, 1990). However, economic constraints have increasingly led
many local councils in Britain to exclude all the above categories
except street living from the definition of homelessness (Gay, 1989).
Because of some of these problems with definitions, precise comparisons of the number of homeless in London and NYC are difficult.
152
Theoretical approach
A cross-national comparison provides an excellent laboratory for
testing a general model of homelessness proposed by Susser et al.
(1993), and its adaptations for use with older homeless populations
(Cohen, 1996). This model theorizes that various background factors
(e.g. gender, race, parental socio-economic status), disruptive events
in youth and adulthood experiences (e.g. occupational history,
physical and psychiatric status, deviant behaviour, social supports)
may predispose persons to homelessness. These predisposing factors
interact with structural forces such as housing and economic conditions. Finally, there may be immediate precipitating events that tip a
person into homelessness. On becoming homeless, a person may
quickly find housing, remain chronically homeless, or alternate
between homelessness and domiciled status. A combination of individual and structural forces affect the ability of an individual to
become successfully re-domiciled.
153
154
155
care or institutional living in their youth, and about two thirds were
raised by both parents. Although there are no data regarding foster or
institutional care among London's ageing homeless, Marshall &
Reed (1992) reported that nearly one fifth of their sample had
experienced significant separations from one or both parents and
16% reported an alcoholic parent.
Deviant behaviour
It is difficult to gauge the meaning of criminal arrests among ageing
homeless persons because many were arrested for loitering, vagrancy,
public intoxication and the like. There were considerable similarities
in history of incarceration between the two cities. In NYC, history of
arrests were 49% and 15% for men and women, respectively (Cohen,
1996). In London, arrest rates were 52% for men in Weller et fl/.'s
(1989) one-night surveys between 1985 and 1988 and 30% for
women in direct access hostels (Marshall & Reed, 1992).
Alcohol abuse was a common problem among the homeless in both
cities, with rates among older men being considerably higher than
among older women. Currently, about two fifths of the men in
Cohen's study in NYC and Crane's (1993) study in London were
problem drinkers. A survey of older residents at Arlington House
(Wake, 1991), a large direct access hostel for men, found one third
were heavily 'dependent' on alcohol. However, Timms & Fry (1989)
found only 3% of older direct access hostel residents to be clinically
diagnosable as alcohol-dependent. Among homeless women the
prevalence rate was 8% in NYC (Cohen, 1996) and 0-16% in various
London sites (Marshall & Reed, 1992; Crane, 1993). However, in
both cities, as many as one third of women may have life-time
histories of alcoholism (Marshall & Reed, 1992; Cohen, 1996).
History of substance abuse was less than 5% in both cities (Ladner,
1992; Marshall & Reed, 1992; Cohen, 1996).
Mental health
Mental illness has been identified as a major contributory factor to
homelessness in both the USA and Britain. Between 20% and 70% of
London's older homeless have been found to have active psychotic
symptoms, 'thought disturbances', a diagnosis of schizophrenia, or
prior psychiatric history (Timms & Fry, 1989; Weller et al., 1989;
156
Marshall & Reed, 1992; Crane, 1994). Rates for women were about
twice that of men. These figures were remarkably similar to those of
NYC's ageing homeless. For example, Cohen (1996) found 54% of
women and 23% of men exhibited psychiatric symptoms or had prior
psychiatric hospitalization; and among shelter residents aged 50 and
over, Crystal (1984) found 27% of women and 12% of men had prior
psychiatric hospitalization.
Prevalence of depression was also very high in both cities. Two
fifths of London's elderly street homeless and one third of NYC's older
street and shelter homeless reported substantial levels of depressive
symptoms (Crane, 1994; Cohen, 1996). This widespread despair was
illustrated by the fact that 30% of NYC's older homeless agreed that
'life was not worth living'. Whether these symptoms correspond to
actual clinical diagnosis is unclear. For example, the prevalence of
diagnosable affective disorders was less than 2% among London's
older hostel dwellers (Timms & Fry, 1989; Marshall & Reed, 1992).
Levels of cognitive deficits differed between the two cities. Moderate or severe cognitive disturbance was found in 39% of London's
street homeless (Crane, 1994), approximately half of whom were
judged to have severe problems. However, fewer than 4% of older
hostel dwellers had clinical diagnoses of organicity and dementia
(Timms & Fry, 1989; Marshall & Reed, 1992). In NYC, 9% of street,
shelter, or flophouse dwellers had moderate or severe deficits, and one
third of these were judged to be severe (Cohen, 1996).
Physical health
In both cities, the ageing homeless were thought to have impaired
physical health. Many suffered from the so-called 'Skid Row Syndrome' of respiratory disease, gastro-intestinal disease, hypertension,
seizures and physical trauma (Feldman etal., 1974; Ashley etal., 1976;
Timms & Fry, 1989). Among London's street homeless (Crane,
1993), 17% of the sample had mobility problems whereas 30% of the
NYC street and shelter sample (Cohen, 1996) had ambulatory
difficulties. Weller et al. (1989) reported that two fifths of London's
homeless persons required medical treatment. Many London hostels
housed very impaired older persons. For example, one hostel reported
(Wake, 1991) that one fifth of its tenants were unable to walk three to
four steps without help or were dependent on a wheelchair, and two
fifths had difficulty with stairs.
157
158
(Cohen, 1996). Moreover, older NYC homeless men and women had
mean social networks of 8.5 and 11.4 persons, respectively.
Patterns and pathways of homelessness
There was little evidence in either NYC or London that older
homeless persons were largely dischargees from institutions. Fewer
than 12% had been in institutions prior to becoming homeless
(Marshall & Reed, 1992; Crane, 1993; Cohen, 1996). Among those in
shelters or streets in NYC and London (Crane, 1993; Cohen, 1996)
about two thirds had reported living in apartments or houses and
about one quarter had been living in SRO hotels, hostels, or boarding
homes prior to becoming homeless. In NYC, half reported living
independently and two fifths said they had been living with family or
friends (Ladner, 1992; Cohen, 1996).
Although many persons stated that they had drifted into homelessness, in some instances the first episode of homelessness was triggered
by a specific event. The factors identified by self-report as more
immediate causes of homelessness were similar in NYC and London
(Ladner, 1992; Crane, 1993; Cohen, 1996): (a) financial, e.g. loss of
welfare support when children or grandchildren moved, rent raised,
loss of work; (b) breakdown of relationships due to death of spouse or
kin, conflict, or abuse; (c) eviction, sale of property, fire, or unsafe
conditions; (d) dissatisfaction with residence. Observations by researchers in both cities suggested that alcoholism and mental health
problems such as paranoid ideation also appeared to contribute to
creating homelessness (Marshall & Reed, 1992; Crane, 1993, 1994;
Cohen, 1996). On the basis of a variety of factors such as life-style,
family support and availability of public welfare, women are much
less apt to become homeless (Marin, 1991). Moreover, in NYC,
(Cohen, 1996), there was approximately a 10 year gap in the age that
women first became homeless (early fifties) versus men (early forties).
Those older persons in both cities who lived in the lowest level of
accommodations such as hostels or flophouses were able to develop a
greater level of stability than those in shelters or streets. For example,
the median length of time in flophouses or hostels was 2 years whereas
women in NYC shelters had a median length of stay of 4 months
(Marshall & Reed, 1992; Cohen, 1996). Among the older women in
NYC shelters, three common patterns were discerned: (a) prolonged
homelessness of 2 or more years (37%); (b) usually domiciled except
159
160
161
generally remain in contact with the family, even if they are living in a
shelter. In Britain, on the other hand, because separation from the
family is rare, when it occurs, it usually reflects a severe rift, sometimes
triggered by an older person's psychopathology. Consequently, the
older homeless in London report few contacts with kin and many
years of isolation.
There are several structural factors that affected older British and
American homeless alike (Hopper, 1988; Greve, 1991). Both Britain
and the USA experienced economic downturns in the early 1980s.
Accompanying this recession was the permanent loss of manufacturing jobs as London and NYC shifted increasingly to service economies. Many persons, especially those who were older and blue collar
workers, were unable to re-enter the workforce during the economic
recovery that ensued.
Secondly, both Britain and the USA have undergone large-scale
release of patients from psychiatric hospitals, although the pace in
Britain has lagged behind that of the USA. Thus, there was an 80%
drop of psychiatric inpatients in the USA between 1955 and 1990,
whereas the British experienced a 60% decrease over the same time
period (Murphy, 1991). In both countries many older inpatients
were 'trans-institutionalized' into nursing homes, although this
practice has been curtailed in both countries. In Britain, for example,
more older inpatients are being discharged to residential care
programmes (Murphy, 1991). Nevertheless in both countries, there is
an inadequate array and supply of residential treatment programmes
that cater to older adults (Lipmann, 1995).
A third important structural factor concerns access to benefits,
health care and housing. Thus, whereas all Britons aged 60 and over
are legally entitled to sufficient income to live above the poverty
level, to health and psychiatric care under the National Health
Service and to priority placement for permanent housing, there is
evidence that the poorest segments of the elderly population do not
always receive these benefits. For example, with respect to housing,
local authorities had varying interpretations of age vulnerability;
they often considered any 'reasonable' offer of permanent housing
to fulfill their obligations; interpretations of homelessness differed
with only 30-40% of authorities considering persons in bed and
breakfast hotels or hostels as homeless; and many authorities were
often insensitive to older persons and perceived them as 'lost causes'
(Gay, 1989; Kelling, 1991). Weller et al.'s (1989) survey found that
162
163
Solutions
On a broader societal level, there are important differences in how
the British and Americans must approach the problem of ageing
homeless. In Britain, in theory there are a variety of statutory social
programmes that should provide adequate income support, health
care and housing. There are less firm supports for psychiatric
aftercare, but this has begun to be addressed under the new legislation
(Murphy, 1991). In the USA, considerably less statutory support is
available to older persons - especially for those between the ages of 50
and 64 - and what is available may not be easily accessed. Thus, we
164
Table 10.1. Older homeless and community samples in New York City
and London: individual characteristics
Physical disorders
Activity limitations
Psychosis/prior
psychiatric
hospitalizationb
Depression
Cognitive deficits
(moderate/severe)
Low social contacts
Marital status
Never married
Divorced/Separated
Widowed
Never had children
Alcohol usec
Poor grooming/
hygiene0
Low skilled
occupations
Racial minorities
Disruptive youth
History of
imprisonment
London
homeless
+++
NA
+++
+++
++
+++
+
++
+
+++
+ / + + +d
+++
+
+
+
+++
++
+
+
+
+
+
+
++
+
++
++
+++
4++
+++
-f + +
+
+
+++
+
+
+
+++
+++
+
++
+++
+++
++
+++
++
+
+
++
+
Note: For each variable, the lowest score is given a value of ( + ) and the other
scores are approximate multiples of this baseline score.
a
Gurland et al. (1983).
b
Women > men.
c
Men > women.
d
Street homeless.
165
166
Summary
A comparison of older homeless in London and NYC indicated the
following:
1. Although personal biography and traits contribute to individual
homelessness, structural forces largely determine the extent of
homelessness and, to a lesser degree, the patterns of homelessness.
2. Statutory regulations are necessary but not sufficient to provide
housing and income support to older persons. In difficult economic periods, governments may discourage persons from seeking
assistance, and strong advocacy is required to secure entitlements.
3. Most older homeless persons in both cities do not prefer to live on
streets or in shelters. However, any housing scheme must recognize
the desire for independence expressed by many older homeless
persons as well as the high levels of physical and mental illness both psychoses and cognitive disorders - that may necessitate
more intensive supervision and care. Thus, any solution must
entail a suitable mix of age-segregated, safe and supportive
housing programmes.
Acknowledgements
The authors greatly appreciate the assistance of Heather Petch, Mary
Carter, Jay Sokolovsky, John Mason, Kim Hopper, Frank Lipton,
Diane Sonde, Ellen Baxter, Eric Roth, Eugene Feigelson and Carole
Lefkowitz. This work was supported in part by NIMH Center for
Mental Disorders Branch grant no. RO1-MH45780.
167
References
Access to Health (n.d.) Health Problems of Homeless People. London: Access
to Health.
Adams, G. A. (1986). Homelessness in the postindustrial city. Urban
Affairs Quarterly, 21, 527-49.
Ashley, M. J., Olin, J. S., le Riche, W. H., Kornaczewski, A., Schmidt,
W. & Rankin, J. G. (1976). Skid row alcoholism: a distinct
sociomedical entity. Archives of Internal Medicine, 136, 272-8.
Bachrach, L. (1987). Homeless women: a context for health planning.
The Milbank Quarterly, 65, 371-96.
Bogue, D. J. (1963). Skid Row in American Cities. Chicago: University of
Chicago Press.
Camden Welfare Rights Unit (1992). Report on the Benefits and Mental
Health Project. London: Corporate Services.
Coalition for the Homeless (1984). Crowded Out: Homelessness and the
Elderly Poor. New York: Coalition for the Homeless.
Cohen, C. I. (1996). The elderly homeless: a conceptual model. In P.
Szwabo & G. T. Grossberg (eds.), Psychosocial Needs of Special
Populations of the Elderly, (in press).
Cohen, C. I. & Sokolovsky, J. (1989). Old Men of the Bowery. New York:
Guilford.
Cohen, N. L. & Sullivan, A. M. (1990). Strategies of intervention and
service coordination by mobile outreach teams. In N. L. Cohen
(ed.), Psychiatry Takes to the Streets, pp. 63-79. New York: Guilford
Press.
Cohen, C. I., Onserud, H. & Monaco, C. (1993). Outcomes for the
mentally ill in a program for older homeless persons. Hospital and
Community Psychiatry, 44, 650-6.
Cohen, C. I., Sokolovsky, J., Teresi, J. & Holmes, D. (1988). Gender,
networks, and adaptation among an inner-city population. Journal
of Aging Studies, 2, 45-56.
Crane, M. (1990). Elderly Homeless People in Central London. London: Age
Concern England and Age Concern Greater London.
Crane, M. (1993). Elderly Homeless People Sleeping on the Streets in Inner
London: An Exploratory Study. London: Age Concern Institute of
Gerontology.
Crane, M. (1994). The mental health problems of elderly people living
on London's streets. International Journal of Geriatric Psychiatry, 9,
87-95.
Crystal, S. (1984). Homeless men and homeless women: the gender gap.
Urban Social Change Review, 17, 2-6.
Department of Environment (1993). Quarterly Statistics. December, 1993.
London: HMSO.
Federal Task Force on Homelessness and Severe Mental Illness (1992).
Outcasts on Main Street. Washington, DC: Interagency Council on
the Homeless.
168
Feldman, J., Su, W. H., Kaley, M. & Kissin, B. (1974). Skid row
and inner-city alcoholics. A comparison of drinking patterns
and medical problems. Quarterly Journal of Studies on Alcohol, 35,
565-76.
Gay, O. (1989). Homeless. Background Paper No. 229. London: House of
Commons Library Research Division, 4 July 1989.
Gay, O. & Greener, K. (1990). The young single homeless. Background
Paper No. 242. London: House of Commons Library Research
Division, 9 March 1990.
Gelberg, L., Linn, L. S. & Mayer-Oakes, S. A. (1990). Differences in
health status between older and younger homeless adults. Journal of
the American Geriatrics Society, 38, 1220-9.
Greve, J. (1991). Homelessness in Britain. York: Joseph Rowentree
Foundation.
Gurland, B., Copeland, J., Kuriansky, J., Kelleher, M., Sharpe, L. &
Dean, L. L. (1983). The Mind and Mood of Aging. New York:
Haworth Press.
Havesi, D. (1991). Census count of homeless is disputed. New York
Times, 13 April, p. 26.
Hopper, K. (1988). More than passing strange: homelessness and mental
illness in New York City. American Ethnologist, 15, 155-67.
Hopper, K. (1991). Homelessness old and new: the matter of definition.
Housing Policy Debate, 2, 757-813.
Interagency Council on the Homeless (1991). Executive Summary. The 1990
Annual Report. Washington, DC: Interagency Council on the
Homeless.
Keigher, S. M. (ed.) (1991). Housing Risks and Homelessness Among the
Urban Elderly. New York: Haworth Press.
Kelling, K. (1991). Older Homeless People in London. London: Age
Concern.
Kelling, K. (1992). Meeting the health needs of older homeless people.
Health Visitor, 10, 346-7.
Ladner, S. (1992). The elderly homeless. In M. J. Robertson & M.
Greenblatt (eds), Homelessness: A National Perspective, pp. 221-6. New
York: Plenum Press.
Lipmann, B. (1995). The Elderly Homeless. Williamstown, Australia:
Wintringham Hostels.
Marcuse, P. (1990). Homelessness and housing policy. In C. L. M.
Caton (ed.), Homeless in America, pp. 138-59. New York: Oxford
University Press.
Marin, P. (1991). Why are the homeless mainly single men. The Nation,
253, 46-51.
Marshall, E. J. & Reed, J. L. (1992). Psychiatric morbidity in homeless
women. British Journal of Psychiatry, 160, 761-8.
Martin, M. A. (1982). Strategies of Adaptation: Coping Patterns of the
Urban Transient Female. PhD thesis, Columbia University, NY.
Murphy, E. (1991). After the Asylums. London: Faber and Faber.
169
New York City Commission on the Homeless (1992). The Way Home. A
New Direction in Social Policy. New York: Office of the Mayor.
New York State Coalition for the Homeless (1989). One Hundred
Thousand, and counting ... Homelessness in New York State. Albany, NY:
New York State Coalition for the Homeless.
O'Flaherty, B. (1991). What's homelessness? Comparing New York and
London. Paper presented at the 13th Annual Conference of the
Association for Public Policy Analysis and Management, Bethesda,
MD, 24-26 October.
Piachaud, D. (1992). Hopes that turned to dust. The Guardian, 30
September, p. 21.
Primary Care for Homeless People Team (1991). Annual Report 1991.
London: Camden & Islington Family Health Services Authority.
Schmidt, W. E. (1992). Across Europe, face of homeless become more
visible and vexing. New York Times, 5 January, pp. 1, 8.
Scott, J. (1993). Homelessness and mental illness. British Journal of
Psychiatry, 162, 314-24.
Scott, R., Gaskell, P. G. & Morrell, D. C. (1966). Patients who reside in
common lodging houses. British Medical Journal, 2, 1561-4.
Susser, E., Moore, R. & Link, B. (1993). Risk factors for homelessness.
American Journal of Epidemiology, 15, 546-56.
Tiadail, S. (1992). U.S. poor get poorer. The Guardian, 5 September,
p. 9.
Timms, P. W. & Fry, A. H. (1989). Homelessness and mental illness.
Health Trends, 21, 70-1.
Toro, P. & Rojansky, A. (1990). Homelessness: some thoughts from an
international perspective. Community Psychologist, 24, 8-11.
Wallace, S. E. (1968). The road to skid row. Social Problems, 16, 92-105.
Wake, M. (ed.) (1991). Housing and Care Needs of Older Homeless People.
London: Arlington House.
Weeden, J. & Hall, C. (1985). The Homeless Mentally III Elderly.
University of California, S.E.: Institute for Health and Aging.
Weller, M., Tobiansky, R., Hollander, D. & Ibrahami, S. (1989).
Psychosis and destitution at Christmas 1985-88. Lancet, ii, 1509-11.
11
Primary health care of the single homeless
DAVID E L - K A B I R AND SIMON RAMSDEN
To write about the primary care of the single homeless suggests that
there are particular factors in that population that are in some way
different to the general population, and that these influence care. It is
important to understand these factors if we are to look at primary care
in this setting. To start with, who exactly do we mean and can we
define the population? This is far from straightforward but definitions
should include inadequate accommodation and some degree of
disaffiliation from society. 'Single' homeless usually means those who
are unmarried or divorced and without close supportive relationships, but occasionally married couples are seen within groups
considered to be single homeless. In this chapter, we will look at the
care of a wide and disparate population, living on the streets, in
hostels, night shelters and bed and breakfast accommodation, as well
as those in more permanent accommodation but who have previously
had more tenuous accommodation.
A morbidity profile
A number of studies have described the physical and psychiatric
morbidity of a variety of single homeless populations. In terms of
physical health, frequent infectious, respiratory and skin diseases and
infestations are usually described, along with trauma (Hewetson,
1975; Powell, 1987; Shanks, 1988). Within this perhaps obvious
pattern are more serious diseases, less prevalent in the general
population. Tuberculosis is the best known (Patel, 1985) although
others such as epilepsy are found more frequently than expected.
Over 50% of hostel populations are usually found to suffer from a
170
171
600 r
Figure ILL Presenting problems of new patients. Note that 40% of new patients
presented with psychiatric problems.
chronic disease (Scott et aL, 1966). Studies of psychiatric disorders
show wide variations, schizophrenia is common and its prevalence
has been estimated in one study as over 40% (Weller et aL, 1989).
Personality disorder, a term not usually defined precisely, has been
described at up to 50% in some studies (Lodge Patch, 1971). Major
affective disorders are often found to be more common than expected.
Alcoholism is common, as is drug abuse, and often co-exists with
severe psychiatric or physical illness. Figure 11.1 shows the morbidity
pattern of presenting patients at our medical centre in 1992-1993,
and is typical of that recorded amongst homeless populations.
The variation in morbidity patterns reported in various studies is
wide, and in large part reflects the different settings in which the
studies were conducted. This has important implications as the single
homeless are often erroneously thought to be an essentially homogeneous group by politicians and health planners. The studies
showing a high prevalence of schizophrenia, usually with much lower
levels of alcohol abuse, have generally been conducted in hostels or in
ill-defined groups such as those attending the Crisis at Christmas
emergency shelter. Street dwellers are more difficult to assess, but in
our study of those staying at a cold weather shelter in London at a
time of deep snow, the pattern was predominantly that of alcohol
abuse (over 60%) and only 9% were defined as definitely or probably
172
psychotic (Reed et al., 1992). Whilst the shelter stayed open, the
number of alcoholics fell rapidly due to aggressive and disruptive
behaviour, whilst the number of schizophrenics rose as they were far
easier to cope with. Studies of young homeless people find less severe
mental illness but high levels of anxiety, family problems and drug
abuse. Overall, around 80-90% of single homeless people are male,
although the sex balance is far more even in the younger groups.
There are few studies of older homeless women, but what data are
available suggest a high prevalence of severe mental illness (Marshall
& Reed, 1992). A number of classifications of homelessness have been
described, perhaps the most useful of which is that of Fischer &
Breakey (1986). They recognized four groups: the chronically mentally ill, alcohol abusers, the situationally distressed and street
dwellers. These groups vary in terms of morbidity, age structure and
life-styles. The young (age < 20 years) are more likely to suffer
situational distress, the chronically mentally ill are more likely to
reside in hostels as described, and alcohol abusers on the streets, and
both of these groups tend to be older. One way to look at the
morbidity patterns is to distinguish between those illnesses that are a
consequence of homelessness, and those which may result in homelessness. Surprisingly, the pathways of social, psychological and
medical problems underlying homelessness have received little attention as yet. There is some evidence that schizophrenia precedes the
onset of homelessness, and that the mental, family and social
deterioration that the illness causes underlies the subsequent homelessness (Tidmarsh & Woods, 1972). Similar evidence is available for
homeless people who abuse alcohol (Myerson & Mayer, 1966), and
recently it has been suggested that the social and psychiatric problems
that are more common in children with epilepsy predisposes to
subsequent homelessness, explaining to an extent the excess of this
condition seen among the homeless (Ramsden, 1994). It has been
considered that many of those who previously were in long-stay
psychiatric hospitals would subsequently become homeless although
in contrast to the USA, there is little evidence as yet that this is the
case (Geddes et al., 1994). Tuberculosis in this context is a result of
homelessness: one hostel study from the USA showed that the
likelihood of a positive tuberculin test increased with the duration of
stay at the hostel (McAdam et aL, 1985).
173
174
this are that whilst some of the external barriers to consultation have
been addressed, the internal barriers particularly emotional ones are
ignored. Many patients express guilt at their circumstances and
life-style, and others an expectation that they will be unwell. Some
find sitting in waiting rooms with patients who are not homeless
threatening and humiliating. Nonetheless illness is frequent and may
necessitate consulting a doctor. 'Flu' or leg ulcers or injuries may
make coping with the demands of homelessness difficult, and other
symptoms such as cough or abdominal pain may be understood as
evidence of a life-threatening disease, that the individual feels is the
result and perhaps desserts of social failure and homelessness.
175
The first point of contact is the most crucial one, and its model
applies to all subsequent encounters. The role of the receptionist is
therefore, in this respect, crucial in setting the ambience of the
consultation. The aim is the same for all the care workers: to make
manifest one's intentionality, the basic motive behind one's choice of
action. One's perception of another's situation leads to a number of
different options, to which the doctor may not be sensitive, owing to
his own past, or other circumstances of which he may not be entirely
aware. To do that implies a degree of self-awareness that requires
training and development. One has to put across one's humanity
rather than one's sympathy, and also one's desire to understand. A
thorough professionalism and an attention to detail, especially to
shades of meaning and to fleeting impressions, is essential. Someone we wish we could remember who wrote: Ha verite est dans les nuances'*
(truth is in the nuances). How very true. There is no room here for
cosmetics. It is not much use maintaining eye contact if this is
perceived by the other person to reveal a glazed look and an
invitation to collusion. The doctor has a need to feel effective, as much
as the patient has a desire to be reassured. This may lead to short-term
solutions and to superficial insights, a sort of psychological treading
water. Collusion and care can thus become mutually exclusive. The
role of rapport with homeless people has been illustrated by Shanks
(1981). He interviewed a group of his patients obtaining sociodemographic information, this was repeated 6 months later. The
discrepancy in the data was less than 1%. A colleague interviewed a
different group, introducing himself as a hospital doctor doing
research. His data showed over a third of data items were discrepant.
Clearly the responsibility for building up this trust is heavily
weighted towards the doctor. He has more to do, more adjusting of
attitudes to perform than the patient, who is a supplicant with
problems that are both complex and obscure. To understand these
the doctor has not merely to be on top of his metier, but he has to
evaluate the situation that faces him and gauge the appropriateness of
his response to it. It is our belief that people who are deprived or
oppressed are generally very observant and astute as to the personality of the people they deal with. An eminent chest physician who has
been interested in some aspects of the care of the homeless recently
mentioned to us that when he was interviewing homeless patients, he
felt that he was himself being interviewed! It has to be added that this
evaluation is all the more potent because it is invariably wordless.
How is the doctor to survive this appraisal and how is he to establish
176
111
178
best when there is liaison between the centre staff and the medical
services, and where the support of trustworthy staff can encourage the
use of the services by reluctant people. We have found that patients
will often develop a considerable degree of trust in the individual
clinics, although they would still be cautious of seeing other doctors,
particularly in a hospital setting. Some claim that such services
increase the marginalization of the homeless by restricting their
access to 'mainstream' services, although those who argue this rarely
attempt to address the problems described above. Specialist clinics
have evolved to care for the homeless and can function as a point of
re-entry to services and society as a whole. Some have proved
remarkably successful in gaining the trust of both patients and
homeless networks, and have been able to orientate their services and
style to their needs. For example, Great Chapel Street Medical
Centre in Soho, London has open access to all services, including
psychiatry as well as offering nurse, chiropody, dental and social
work care. The centres can also gain expertise in the best use of social
workers, housing and other resources, which may be needed to ensure
effective care. They are probably most effective where there are
particularly large populations of homeless people, including those
sleeping rough. These conditions are most likely to be found in large
inner cities. Outreach services can meet a further group of homeless
people who rarely use day centres, hostels or other services. One study
from a mobile clinic that visited various sites in London found that a
third claimed to rarely use other services (Ramsden et al., 1989). Once
trust had been established, people started to use the services, and a
third of those visited started to use a nearby clinic for the homeless
staffed by the same doctors within a month of the first consultation at
the mobile surgery. Outreach services are most effective when the
trust and rapport can be developed as a result of their operation
effectively encourages patients to use services, as there are obvious
limitations to the scope of medical care provided on the streets.
Homeless people often consult with a daunting combination of
physical and psychiatric disorders, and on a number of occasions we
have found that treatment is particularly difficult whilst the patient
remains on the streets. Examples include leg ulcers, pulmonary
tuberculosis, psychosis and severe 'flu'-like illnesses. Such patients are
frequently reluctant to be admitted to hospitals, and in some cases, the
hospitals themselves are reluctant to admit patients when 'social
problems' form such a large component of the illness. In 1984 we
Failure of projects
Whilst the above describes many initiatives and areas to develop, a
relatively large number of projects over the past 20 years have ceased
to exist or failed to develop beyond a quite limited scope. A number of
themes underlie the success or otherwise of projects. Failure may
result from a lack of clarity about the aims of the project or non-viable
goals. These projects have suffered through poor groundwork to
establish what resources were needed and have led to the service not
being used by the homeless. Staff may have unrealistic expectations of
what can be achieved, and rapidly lose heart. Some start or become
isolated from others working in primary care and again become
179
180
References
Balint, M. (1964). The Doctor, his Patient, and the Illness. 2nd edit.
Geddes, J., Newton, R.5 Young, G., Bailey, S., Freeman, C. & Priest, R.
(1994). Comparison of prevalence of schizophrenia among hostels
for homeless people in 1966 and 1992. British Medical Journal, 308,
816-19.
181
182
183
PART III
INTERNATIONAL
PERSPECTIVE
12
European Perspectives
Introduction to European chapter
M A X MARSHALL
188
M. MARSHALL
with children are much less likely to surfer from severe mental
disorder than single homeless women. Hence it is difficult to make
comparisons between US and UK studies of the rates of mental
disorder in homeless UK women as the studies draw on very different
populations.
The main reason for many specious USAUK comparisons is
similarity of language rather than similarity of culture and social
policy. On these grounds a plausible case can be made that European
studies are of equal or greater relevance than USA studies. Yet UK
commentators have shown remarkably little interest in research
carried out by our European partners. This chapter attempts to
redress the balance by presenting the findings of relevant European
research that has remained largely uncited in English language
reviews. The amount of information presented is small in relation to
that which remains to be uncovered (for English readers). Nevertheless, the limited findings presented here demonstrate both the quality
of the research available, and the potential for furthering our
understanding of the causes of homelessness by cross-European
comparisons.
From Denmark, Brandt and Munk-Jorgensen, present a series of
studies of high epidemiological quality that have not previously been
discussed in the English language literature. From a UK point of view
the research has two interesting messages. First, even in societies with
highly developed welfare systems we see the problems of: homelessness amongst the young; accumulation of severely mentally ill in
lodging houses; and direct discharge of psychiatric patients from
hospitals to hostels. Secondly, unlike the UK, Denmark has not seen a
recent rise in the total numbers of homeless people, perhaps because
the state remains committed to providing low-cost accommodation to
those in need.
From Germany, Rossler and Salize again demonstrate that there is
a substantial body of relevant research that has not penetrated the
literature in English. As the authors point out, the interpretation of
the German findings is complicated by two special factors: the
problems of reunification, and the long-term effects of the Nazi policy
of exterminating people with severe mental disorders. Despite these
complicating factors, and the limitations of the available data, there is
evidence that Germany has seen an accumulation of people with
severe mental disorders in hostels for the homeless. It may be that
189
References
Scott, J. (1993). Homelessness and mental illness. British Journal of
Psychiatry, 162, 314-25.
Homelessness in Denmark
PREBEN BRANDT AND POVL MUNK-JORGENSEN
190
191
Men
(%)
Women
(%)
6
94
22
78
100
100
192
193
194
195
very similar to what is reported from bigger cities (Isaac & Armat,
1990), and there will always be about 20-30 persons with severe
mental illness living under very bad conditions, i.e. with dirty clothes,
eating old food from their bags or from litter bins, sitting on the
pavement in all kinds of weather.
In Denmark a debate about these mentally ill homeless people's
right to live as they want has only just been opened. What we
discussed is on the one hand the civil rights of the single homeless
mentally ill and on the other hand the community's responsibility for
ensuring everybody has any necessary treatment and a satisfactory
social standard of living.
Since the beginning of 1992, a programme appealing to the
homeless and homeless-threatened who are severely mentally ill and
who reject treatment has been established in Copenhagen (P. Brandt,
unpublished data). In the first year the programme was in contact
with 37 persons and of these 19 had clearly got a better way of living,
due to the programme.
It is obvious that what these people need is neither simple
sheltering nor ordinary psychiatric treatment. It is also clear that the
homeless psychotics who live in the streets will not go anywhere to ask
for help and they will even usually start by refusing to accept any help
offered.
Therefore, the programme is based on field-work in the streets and
includes a wide spectrum of ways to act, including use of the mental
civil law.
196
References
Amtsradsforeningen i Danmark (The Association of County Councils in
Denmark) (1990). Amterne og videreudviklingen af 105-institutionerne.
(The counties and the development of the Section 105-institutions).
197
Introduction
In Germany, psychiatrists have historically taken an interest in the
origins of homelessness. Early in this century they tended to ascribe
psychopathological reasons for homelessness. In fact, two of the most
famous German psychiatrists of this time, Wilmanns (1906) who
developed the concept of 'Wandertrieb' (migration instinct) and
Schneider (1934) who came out with his theory of psychopathic
personality, were of this opinion. Unfortunately their work was used
to legitimize the murder of thousands of homeless people during the
Nazi regime. It was with this that the Nazis transformed the German
mental health movement, which had it's heyday between the two
wars (Haselbeck, 1985) into a massacre of the mentally disabled.
However, theories of endogenous aetiologies of homelessness managed
to survive the war (Ritzel, 1974; Garcia, 1986). An increased rate of
pathological change in the brains of deceased homeless persons led
Veith & Schwindt (1976) to hypothesize that homelessness was a
symptom of psycho-organic syndromes. This concept was discussed
very controversially (Locher, 1990). The Bodelschwingh institution
in Bielefeld-Bethel, a well established centre for helping the homeless
in Germany, conducted a large interdisciplinary study on aetiology
and phenomenology of homelessness in 1977. Although the scientific
design of this study had promised a profound empirical contribution
to the debate on endogenous versus social factors in homelessness,
there was unfortunately neither interdisciplinary synopsis nor inter-
198
pretation of the results beyond the level of epidemiological statements. Sociological and medical findings were published separately
(e.g. Goschler, 1983; Sperling, 1985), partly differing in description
and interpretation (Locher, 1990).
Nevertheless at the present time the phenomenon of homelessness
in the Federal Republic of Germany is mainly considered to be the
result of an interaction between poverty, unemployment and
social disintegration. Consequently, the old label 'NichtseBhafte'
(vagrants) is gradually being replaced by the label 'alleinstehende
Wohnsitzlose' (single homeless). According to this concept uprooting, homelessness and vagrancy occur increasingly in times of
economic crises, triggered by growing unemployment, a deficient
housing market and an insufficient welfare system (Deutscher Verein
fur offentliche und private Fiirsorge, 1990). Thus, mental disorders
are currently seen as one factor in a number of interacting variables
resulting in homelessness.
One of the main reasons for the sporadic revivals of the discussion
as to whether homelessness is caused by social or psychopathological
reasons is the lack of valid empirical data on the relationship between
homelessness, social distress and mental disorder. Unlike the UK or
the USA, where research on homelessness and mental health is better
established, the issue is still not adequately perceived in Germany.
But the problem has grown in the last few years. As a result of the
German reunification there are great difficulties in reforming the
antiquated mental health care system of the former German Democratic Republic. In the so called 'new' federal states (of the former
German Democratic Republic) large, insufficiently staffed and
equipped psychiatric hospitals are still treating most of the chronically mentally ill. After being discharged, they do not find outpatient
and rehabilitative services of Western standard (Rossler & Salize,
1994). If the communities fail to establish sufficient accommodation
for these patients, they are at a significant risk of being dismissed into
homelessness. Also the immigrants from Eastern Europe (whose
number is steadily rising) run a heightened risk of social disintegration and mental strain with the worsening of the economic situation
and the lack of cheap flats in Germany.
1970 1972
1974
1976
199
1978
1980
1982
1984
1986
1988
Year
Historical development
The widespread neglect of the problem of mentally ill homeless in
Germany in the post-war era was a symptom of the overall neglect of
mental health care during this period. The reform of Germany's
psychiatric care system started as late as the 1970s. One reason for this
delay is the macabre consequence of the Nazi regime, where between
90000 and 120000 mentally ill people were murdered (Finzen,
1983). Therefore the large psychiatric hospitals in Germany did not
suffer from overcrowding to the extent that psychiatric hospitals in
other countries did. With a ratio of 1.6 psychiatric beds per 1000
population during the early 1970s the need for deinstitutionalization
was not as urgent. Nevertheless there was a reduction of beds in
mental health hospitals in Germany of approximately 30% within
the last two decades (see Figure 12.1).
While in countries like the USA many chronically mentally ill were
dismissed more or less directly into homelessness (Dennis et al., 1991),
there was enough accommodation for former long-term patients in
Germany during the first period of deinstitutionalization. Frequently
200
1975
1989
Figure 12.2. Decrease of mean length of stay in psychiatric hospitals in two federal
states of Germany.
, Hessen;
, Rheinland.
201
forced to cut social care offers for the chronically mentally ill. The
hospitals' asylum function disappeared. One indication of this is the
growing number of chronically mentally ill patients referred from
inpatient treatment to social psychiatric mental health centres. In
some areas this rate now amounts to 1520% of all social psychiatric
mental health centres' patients (Rossler et al.> 1993). Already at
earlier stages of the reform in Germany there had been warnings that
deinstitutionalization might become detrimental especially for the
chronically ill (Kunze, 1977).
In the community, mentally ill patients are more exposed to
economic and social stressors than they are behind the protective
walls of the asylum, at least in this case. If the economic situation of a
country gets worse, the mentally ill are among the first to suffer. The
recent German reunification proved this to be true. In the former
German Democratic Republic, many chronically mentally ill had
opportunities to work in state owned enterprises, known as 'Volkseigene Betriebe (VEB)'. Because of the special conditions of Eastern
Germany's economy VEBs could afford to employ a large number of
mentally ill. When these companies were no longer competitive by
the standards of the Western economy, the mentally ill lost their jobs.
Unfortunately, there are no sheltered workshops or other rehabilitative services to which they can go. At the same time the mentally ill
have difficulty finding housing accommodation since rent in Eastern
Germany is rising at an explosive rate.
Current situation
The problem of the mentally ill homeless in Germany can no longer
be concealed. It is acute. However, it is impossible to know the total
number of homeless in Germany. Assessments have to rely on data
from services for the homeless that record contacts. But only a few
services participate. Moreover, there are homeless people who avoid
any contact with care services. Estimations of the size of this group
range from 'very little' to one third of all homeless.
Available data indicate a significant increase in the total number of
homeless during the last few years. At the beginning of the 1970s,
some 26000 homeless persons were registered by the German
Bundesarbeitsgemeinschaft fiir NichtseBhaftenhilfe (Commission for
Care for the Homeless). In 1978 the number increased to 32 000
lime
Population
Klugc (1972)
1968-1971
Psychiatric
morbidity
Alcoholism
Former psychiatric
hospitalization (%)
1.3% psychosis
16.7%
5.2
24%
40% alcoholism
44% at risk
46%
56% alcoholism
23.9% alcohol abuse
3.6
34.7% alcoholism
32.2% at risk
70%
Many
30%
60%
67%
(5-10%
multi-morbidity)
63.5%
36.5
17.5% alcoholism
25% alcohol abuse
5% drug addiction
203
204
205
206
Conclusions
The problem of mentally ill homeless people in Germany is, above all,
a problem of care. The need for mental health care among the
homeless cannot be quantified due to a lack of information. Nevertheless what we do know suggests that this need is considerable. General
services for homeless people carry the main load of mental health care
of their clientele. However, they are not adequately equipped or
staffed for this purpose. They are not able to deliver any form of care
beyond a subsistence level. So these institutions are in most cases
underqualified when they have to deal with people suffering from
mental disorders. The most urgent problem is the alcohol problem.
However, the idea that present services for the homeless in Germany
can substitute for specialized care in the field of alcoholism amounts
only to wishful thinking (Rieger & Wessel, 1992).
Professionals are conscious of this problem, because they experience it daily (Kujat, 1991). However, most new concepts or plans for
the further development of care for the homeless do not take into
account the mental health of this group (LWV, 1993). If reform plans
refer to this problem, the only recommendation is to improve
outpatient care of the homeless in order to ease the burden of
inpatient services for those with alcohol problems (Deutscher Verein
fur offentliche und private Fiirsorge, 1990).
This is not sufficient. Plans that do not consider mental health
needs will not be effective in helping care for the growing number of
homeless people. Data available indicate even now that there is a
substantial need for mental health care for these people.
Exactly what services the mentally ill homeless need and how they
can be combined with the existing mental health care system and the
care system for the homeless cannot yet be determined with our
present knowledge. Further investigations are needed. The results of
international study on this issue will not fill the gap. The historical
development already described in this chapter suggests that the
mentally ill homeless in Germany might differ from those in countries
such as the USA or UK. We can hypothesize that the rate of persons
directly dismissed from inpatient treatment to homelessness in
Germany is smaller. However, there might be a relatively large group
of persons with mental disorders that reject treatment and avoid
utilizing the welfare and health care agencies.
207
References
BMJFFG (Bundesminister fur Jugend, Familie, Frauen und Gesundheit)
(State Department of Youth, Family, Women and Health) (1988).
Empfehlungen der Expertenkommission der Bundesregierung zur Reform der
Versorgung im psychiatrischen und psychotherapeutisch-psychosomatischen
Bereich aufder Grundlage des Modellprogramms Psychiatrie. (Suggestions
made by the Federal Government's Commission of Experts on Reform in the
Psychiatric and Psychotherapeutic-psychosomatic Fields as Based on the Model
Programme of Psychiatry). Bonn: BMJFFG.
Dennis, D. L., Buckner, J. C , Lipton, F. R. & Levine, I. S. (1991). A
decade of research and services for homeless mentally ill persons:
where do we stand? American Psychologist, 46, 112938.
Deutscher Bundestag (1992). Antwort der Bundesregierung auf die Grofie
Anfrage der SPD zur Situation der psychisch Kranken in der fiundesrepublik
Deutschland. (The Federal Government's Response to the Question
Posed by the Social Democratic Party (SDP) Regarding the Situation of the
Mentally III in the German Federal Republic). Bonn: Drucksache
12/4016.
Deutscher Verein fur offentliche und private Fiirsorge (1990). Hilfefur
alleinstehende Wohnungslose (Nichtsefihafte). (Aidfor the Homeless).
Frankfurt/Main: Eigenverlag des Deutschen Vereins fur offentliche
und private Fiirsorge.
Eikelmann, B., Inhester, M. L. & Reker, T. (1992). Psychische
Storungen bei nichtseBhaften Mannern. Defizite in der
psychiatrischen Versorgung. (Mental disorders in homeless men.
Inadequacy in Psychiatric Care). Sozialpsychiatrische Informationen, 2,
29-32.
Finzen, A. (1983). Auf dem Dienstweg. Die Verstrickung einer Anstalt in die
Totung psychisch Kranker. (In the Line of Duty. The Involvement of an
Asylum in the Killing of the Mentally III). Rehburg-Loccum:
Psychiatrie-Verlag.
Garcia, C. (1986). Karl Wilmanns und die Landstreicher. (Karl
Wilmanns and the vagrants). Nervenarzt, 57, 227-32.
Goschler, W. (1983). Die alleinstehenden Wohnungslosen. (Homeless
people). Gefdhrdetenhilfe, 25, 7-14.
Haselbeck, H. (1985). Zur Sozialgeschichte der offenen Irrenfursorge Vom Stadtasyl zum Sozalpsychiatrischen Dienst. (On the social
history of mental health care - the change from urban asylums to
social-psychiatric services). Psychiatrische Praxis, 12, 171-9.
Iben, G. (1989). Armut der Obdachlosen und NichtseBhaften. (Poverty
and the homeless). Blatter der Wohlfahrtspflege, 11/12, 316-20.
Kluge, M. (1972). NichtseBhaftigkeit aus der Sicht einer
Arbeiterkolonie. (Homelessness from the viewpoint of a worker's
commune). Gefdhrdetenhilde, 14, 5-8.
Kujat, H. (1991). Der "immer wiederkehrende Patient" - aus der Sicht
einer Einrichtung der NichtseBhaftenhilfe. (A service for aiding the
208
Hilfe.
MAGS (Ministerium fur Arbeit, Gesundheit und Sozialordnung
Baden-Wurttemberg) (State Department for Health and Social
Affairs, Baden-Wurttemberg) (1982). Hilfenfur Gefdhrdete und
Nichtsefihafte in Baden-Wiirttemberg. (Services for the Homeless and
Those Threatened with Homelessness in Baden-Wiirttemberg). Stuttgart:
MAGS.
Rieger, J. & Wessel, T. (1992). Ambulante und stationare
NichtseBhaftenhilfe: Endstation Armut - wohnungs- und arbeitslose
Abhangigkeitskranke ohne Chance? (Outpatient and hospital aid
for the homeless: destination poverty. Homeless and unemployed
mentally ill persons suffering from an addiction - have they got a
chance?) In G. Wienberg (ed.), Die vergessene Mehrheit, pp. 159-67.
Bonn: Psychiatrie Verlag.
Ritzel, G. (1974). Entwicklung und gegenwartiger Stand der
psychiatrischen NichtseBhaftenforschung. (The development and
present state of psychiatric research on the homeless). Psychiat. din.,
7, 26-49.
Rossler, W., Hafner, H., Martini, H., an der Heiden, W., Jung, E. &
Loffler, W. (1987). Landesprogramm zur Weiterentwicklung der
aufierstationaren psychiatrischen Versorgung Baden- Wiirttemberg - Analysen,
Konzepte, Erfahrungen. (Programme for the further Development of
Community Mental Health Care in Baden-Wiirttemberg -Analyses, Concepts,
Enke Verlag.
Rossler, W. & Salize, H. J . (1993). Planungsmaterialien fur die psychiatriche
209
29, 112-18.
Schneider, K. (1934). Die psychopathischen Persb'nlichkeiten. (Psychopathic
210
J. FERNANDEZ
ness across various centres. This chapter does not look at the rural
homeless in Ireland.
Domestic perspectives
Identifying the homeless
Following the implementation of the Housing Act (Department for
the Environment, 1988), an assessment of housing need was conducted by the Department of the Environment in 1989, wherein 987
homeless people were enumerated around the country. Disquiet
among voluntary agencies about this serious undercounting
prompted a refinement in methodology during subsequent assessments. Thus, 2751 individuals were found to be homeless nationwide
in 1991; an increase of almost 180% on the previous figure (National
Campaign for the Homeless, 1992) and 2667 individuals nationwide
in 1993 (Department for the Environment, 1993).
The Republic of Ireland is populaced by approximately 3.5 million
individuals (Central Statistics Office, 1993) and of these 5000 are
believed to be homeless of whom 750 are young people (National
Campaign for the Homeless, 1992). However, this finding excludes:
(a) approximately 7400 travellers who move between halting sites
countrywide (Barry & Daly, 1986) and technically have no fixed
abode; (b) over 28 600 approved applicants on our housing lists
(Department for the Environment, 1993) with approximately 32 000
child dependants; (c) over 4500 long-stay psychiatric inpatients
(Health Research Board, 1993) who meet the criteria for homelessness specified in the Housing Act (Department for the Environment,
1988); and (d) a further unquantifiable number of'hidden' homeless,
e.g. single adults working as live-in domestics.
The present group of homeless includes: adult males and females at
a ratio of approximately 4:1, an increasing number of children at a
ratio of almost 1:1, and occasionally entire families.
211
Emigration
Emigration has for long served as a safety valve for Ireland's
unemployment problem. However, given recent economic trends the
world over, the number of people emigrating from the Republic
(expressed as the net migration rate) has changed dramatically from
minus 46 000 individuals in the year ending mid-April 1989 (National Economic and Social Council, 1991) to plus 2000 individuals in the
year ending mid-April 1992 (Dail Eireann, 1993) leading in the latter
instance to an inward movement of migrants and/or returning
emigrants. This has led to a growing demand for accommodation.
Lack of accommodation
The availability of low-cost rented accommodation is now being
reduced by conversion and refurbishment into high-cost units
(Kasinitz, 1984). People displaced from their low-cost accommodation are forced to consider other residential options. The latter
include emergency accommodation in hostels and night shelters,
sharing with friends and relatives, or 'squats' or 'sleeping rough'.
Some homeless mentally ill are known to show either self-injurious or
property-destructive behaviour to gain admission to hospital or to
prison.
212
J. FERNANDEZ
blind, deaf mutes, casuals, unmarried mothers and children (InterDepartmental Committee, 1949). By the late 1960s, provisions were
made to unscramble the needs of many of the disadvantaged groups
in County Homes. In 1968, recommendations called for the redesignation of County Homes as Geriatric Hospitals and casuals were
excluded from concern. The latter were seen to constitute a 'social
problem ... not an appropriate problem for consideration in connection with the care of the aged' (Department of Health, 1968, p. 32).
Conditions for the homeless did not improve following the reorganization of the national health-care system in the 1970s. The McKinsey Report (McKinsey & Co, 1974), which contributed to this
reorganization made no mention of the homeless and responsibility
for this 'social problem' was allocated to neither the Special Hospital
Care Programme nor the Community Care Programme within the
new area Health Boards.
Disinvestment in public housing
Progressive disinvestment in Local Authority housing is another
factor that has compounded the current problem of homelessness.
Thus, 7002 public housing units were completed nationwide in 1984
compared to 1482 units in 1992 (Department for the Environment,
1991, 1993). The government's decision to raise the level of housing
starts to 3500 a year nationwide - beginning in 1993 - does not reflect
the need of the Eastern Planning Region (including Dublin and its
dormitory surrounds), which contains 38% of the national population (Central Statistics Office, 1993).
Poverty, morbidity and the 'underclass'
The inter-relationship between long-term unemployment, poverty
and physical and psychological morbidity has been documented
(Nolan, 1990; Whelan^a/., 1991) and the evidence augurs poorly for
marginalized individuals. Murray (1984) has promulgated the view
that long-term unemployment and poverty impact on the creation in deprived inner city areas - of an 'underclass' characterized by
widespread drug abuse, casual violence, petty crime, illegitimacy,
child neglect, work avoidance, welfare dependence, apparent contempt for conventional values and homelessness.
213
Murray's observations (1984, 1989) were confined to developments in the USA and in the UK. However, related concerns are
reflected in a study conducted by McKeown (1991) in the north inner
city of Dublin, wherein it was noted that: (a) the admission rate of
children into care was four times higher than the national average,
(b) the admission rate to public psychiatric hospitals was three times
the national average, (c) the attendance rate for the treatment of
drug abuse was three times the national average per head of
population and (d) the crime rate was four times the Dublin
Metropolitan Area average and eight times the national average.
Homeless children
The Committee on Reformatory and Industrial Schools (1970)
recommended that monolithic institutions then catering for children
who were being reared 'in care' be replaced by smaller home-style
units, with the result that 1500 institutional places were lost by 1983.
Individuals above the age of 16 years were discouraged from leaving
care facilities under the provisions of the antiquated Children Act of
1908. Notwithstanding this, the visibility of homeless children started
increasing, notably in cities around Ireland, and by 1983 it was felt
that a young homeless class was emerging (National Youth Policy
Committee, 1984). The Children Act of 1908 has been replaced by
the Child Care Act (1991) in which the legal definition of a child has
been raised to 18 years and legal responsibility for the accommodation needs of homeless children is placed on area Health Boards. It is
too early to evaluate the impact of this legislation on youth homelessness but the increasing use of 'bed and breakfast' accommodation is
not felt to represent a legitimate response to the needs of this group.
Surveys conducted around the country in 1987 revealed the
presence of 712 homeless children nationwide below the age of 18
years (Streetwise National Coalition, 1988). Media concern about
the lack of detention facilities for juvenile offenders - many of them
homeless is matched by related concern about the lack of adolescent
psychiatric residential facilities for the same population (Streetwise
National Coalition, 1991). However, very little is known about the
prevalence of psychiatric morbidity among homeless children in the
Republic of Ireland.
214
J. FERNANDEZ
Homeless females
Females are believed to be victims of their unequal position in society
and homeless females epitomize this discrepant situation (Kennedy,
1985). Older homeless females are known to live in direct access
hostels while younger females either stay with friends or remain in
unstable and unsuitable relationships (Austerberry & Watson, 1983)
until a major domestic crisis and/or repeated physical violence force
them to seek help (Bell, 1989).
Homeless women are known to experience high rates of psychiatric
morbidity (Marshall & Reed 1992; Scott 1991; also see Chapter 5).
In recent years the admission rate of homeless females to urban
psychiatric hospitals has increased; against a background of disproportionate under-provision for this group in direct access hostels,
refuges and night shelters (Kelleher et a/., 1992). It is felt that the
number of homeless females identified in psychiatric hospitals in any
given year is an underestimate. Homelessness in females tends to be
'hidden', as many of them are inclined to stay with friends and
relatives and to use their latter address as one of convenience.
Homeless families
Very little is known about the extent or severity of psychological
distress among homeless families in the Republic of Ireland.
215
216
J. FERNANDEZ
A survey in Ireland
A prospective survey was conducted from 1 January through to 31
December 1978 in St. Brendan's Hospital, Dublin. A total of 425
psychiatric admissions (313 male and 112 female) were referred for
review but 169 admissions (106 male and 63 female) were excluded,
as they did not meet specified criteria, or had left hospital before they
could be assessed (Fernandez, 1984, 1985). A total of 256 admissions
(207 male and 49 female) were included in the survey. Relevant
findings have been summarized in Table 12.3, wherein the characteristics of this group, including gender-specific differences are evident.
217
218
J. FERNANDEZ
Males
Females
20.9
62.6
16.5
40.5
28.6
53.6
17.8
37.0
73.9
3.5
2.6
19.1
0.9
89.6
80.0
64.3
21.5
7.1
7.1
Nil
64.3
89.3
60.0
21.8
13.0
50.0
10.7
7.1
3.4
Nil
17.9
0.9
0.9
91.3
14.3
Nil
71.4
0.5-375
24.5
17.4
70.4
1.5-101b
40.5 b
39.3
85.7
90.0
10.0
77.5
22.5
45.0
48.8
6.2
30.6
59.2
10.2
46.0
33.3
7.2
4.3
2.9
1.9
1.9
1.5
4.1
34.7
10.2
Nil
14.3
Nil
Nil
Nil
1.0
36.7
219
Males
Females
55.6
17.4
39.3
10.7
27.0
50.0
0.5-293
9.1
0.5-345
10.5
81.0
12.6
3.0
3.4
Nil
61.2
22.5
8.2
6.1
2.0
1-7
4
1-5
3
60.9
60.7
220
J. FERNANDEZ
Start of programme
325
300
275
250
225
200
1 175
1 150
< 125
100
75
50
Data not
available
Number of patients admitted
25
i
i
i
i
i
III
1975 76 77 7 8 7 9 8 0 8 2 8 2 8 3 8 4 8 5 8 6 8 7 8 8 8 9 9 0 9 2
92
7 6 7 7 7 8 7 9 8 0 8 1 8 2 8 3 84 8 5 8 6 8 7 8 8 8 9 90 9? 92 1993
Years
Figure 12.3. Annual admission rate (homeless men).
Table 12.4. Programme for the homeless (admission unit): Audit of admissions (1979 j 80 to 1992/93)
Years
79/80
1-8
83/84
84/85
85/86
86/87
87/88
88/89
89/90
1-8
1-5
1
1-7
2
1-6
2
1-9
2
1-5
2
1-8
3
1-7
1
1-7
1
Range of admissions
Median re: admissions
Percentage of patients
with one admission
annually
67.96
60.53
79.81 76.77
70.87
67.03
67.07
58.02
67.6
57.83
New contacts
103
+42
+78
+60
+39
+34
+38
+29
+30
1-6
1-6
1
+46
1-6
3
1-9
3
58.21 57.38
60.34
54.09
+33
+22
+16
+19
222
J. FERNANDEZ
Table 12.5. Programme for the homeless: Outcome data (114/79 to 3113/93)
Number of patients
(A) Patients serviced
Through Admission Unit and Day Centre
Through Day Centre with no admission
Total
(B) Known deaths
(C) Re-united with family and/or returned to former
domicile
(Ireland x 8; England & Wales x 9; Italy x 1;
South Africa x 1; India x 1).
(D) Replaced/retained in institutional settings
(E) Placed in community based supervised settings
(F) Placed in the community ( + Day Centre support)
Dublin Corporation flats
Other city flats/bedsitters
Supervised lodgings
(G) Screening for tuberculosis
Tuberculosis discovered in
nursing cover from 8.00 a.m. to 8.30 p.m. each day. Facilities and
services available in the Day Centre include the following: meals,
medication, contact with medical staff, follow-up of other outstanding matters e.g. medical tests, appointments in other city hospitals
and contact with welfare agencies, a 'Follow-up Service' in the
community to track infrequent attenders, a 'Money-Management
Service5 for those who are unable to handle their finances, along with
emergency dockets for accommodation in city hostels, facilities for
bathing and 'launderette' facilities.
223
Outcome data
Table 12.5 summarizes outcome data relating to all individuals who
passed through the two hospital based components of our Programme. Attention is drawn to the high mortality rate of our patients,
i.e. 8.28%, and this refers only to 'known deaths'. Thus, few of the
homeless patients in our Programme are felt likely to achieve
sufficient seniority to avail themselves of psychiatric services for the
elderly.
Data presented at (D), (E) and (F) in Table 12.5 represent a
snapshot of placements at the time these were made. The extreme
mobility of our population, even among those retained in institutional settings, does not guarantee their current presence in the stated
locations. However, all attempts have been made to ensure that
patients were not counted-in more than once in the groups reported.
Incidence of tuberculosis
Results of the National Tuberculosis Survey in the Republic of
Ireland indicate that the incidence rate of tuberculosis in the 26
Counties at 21.37 new cases per 100000 population is the highest in
Western Europe and is twice that of Great Britain and Northern
Ireland (Stinson /#/., 1988; Fogarty, 1991). Data presented at (G) in
Table 12.5 indicate that tuberculosis was detected in 2.34% of our
homeless population. When patients are known to be infected and
infective, provisions exist to ensure their removal, but none to
guarantee that they remain in specialist hospitals for treatment of
their condition, until they are free from infectivity. This poses a
particular problem in the management of subgroups of the homeless
mentally ill, notably those with a history of alcohol and drug
dependence.
High prevalence rates for tuberculosis have also been reported in
the USA among homeless populations; notably among homeless
urban alcoholics. A linear relationship is believed to exist between the
prevalence of tuberculosis and the length of homelessness, with
substance abusers including alcoholics, emerging as the most vulnerable (Brickner et al., 1984).
224
J. FERNANDEZ
225
226
J. FERNANDEZ
decade, it is improbable that initial misperceptions will serendipitously translate into adequate provision.
Conclusion
When Community Care was being proselytized, a theme repeatedly
emphasized was the provision of community based alternatives to
institutional care (Department of Health, 1966, 1984). An equally
important theme, namely the provision of adequate funding to
support proposed developments, was less forcefully presented. The
former theme has now evolved into a major symphonic work
(Murphy, 1991) while the latter is barely audible.
Recent evidence suggests that the percentage of the public health
services budget allocated to psychiatric care in the Republic of
Ireland has fallen from 13.5% in 1971 to 10.1% in 1992 and is
estimated to have dropped further to 9.8% in 1993 (Institute of
Public Administration, 1993). Psychiatrists and their colleagues who
work in the public sector derive comfort from neither the statistical
validation of known misery, nor the curiously confusing commentaries on the funding of services (Editorial, 1990; Walsh, 1994). Public
psychiatric patients have for long been placed on the lowest rung of
the ladder of health-need, and the homeless mentally ill who form a
subset of this group are inevitably left holding the ladder, while their
domiciled and politically more valuable peers are facilitated to
ascend.
References
Austerberry, H. & Watson, S. (1983). Women on the Margins. London:
City University Housing Group.
Barry, J. & Daly, L. (1986). The Travellers' Health Status Study. Dublin:
The Health Research Board.
Bell, J. (1989). Women and Children First. Dublin: National Campaign for
the Homeless.
Birley, J. (1990). Blame homelessness, not the hospitals. The Guardian, 14
March, p. 21.
Brickner, P. W., Filardo, T. Iseman, M., Green, R., Conanan, B. &
Elvy, A. (1984). Medical aspects of homelessness. In The Homeless
Mentally III, H. R. Lamb (ed.), pp. 243-55. Washington, DC:
American Psychiatric Association.
227
228
J. FERNANDEZ
Office.
Marshall, J. E. & Reed, J. L. (1992). Psychiatric morbidity in homeless
women. British Journal of Psychiatry, 160, 761-8.
Murphy, E. (1991). Community mental health services: a vision for the
future. British Medical Journal, 302, 1064-5.
Murray, C. (1984). Losing Ground: American Social Policy 1950-1980. New
York: Basic Books Inc.
Murray, C. (1989). Underclass. The Sunday Times, Magazine Section, 26
November, pp. 26-46.
National Economic and Social Council (1991). The Economic and Social
Implications of Emigration. Dublin: National Economic and Social
Council.
National Youth Policy Committee (1984). Final Report. Dublin:
Stationery Office.
National Campaign for the Homeless (1992). Promises, Promises. An
Assessment of the Effectiveness of the Housing Act 1988 in Housing
229
Supplement 4 , 6 1 .
Statistical Office of the European Communities (1994). Unemployment
in the European Union. In Unemployment, Monthly, March.
Luxembourg: Eurostat.
Stinson, J., Kelly, P., Howell, F. & Clancy, L. (1988). National
tuberculosis survey (1986). Irish Medical Journal, 81, 7-10.
Streetwise National Coalition (1988). A National Survey of Young People
Out of Home in Ireland. Dublin: Streetwise National Coalition.
Streetwise National Coalition (1991). At What Cost? A Research Study on
Residential Care for Children and Adolescents in Ireland. Dublin: Focus
Point.
Talbot, J. A. & Lamb, H. R. (1984). Summary and recommendations.
In H. R. Lamb (ed.), The Homeless Mentally III, pp. 1-10.
Washington, DC: American Psychiatric Association.
Walsh, D. (1994). Letters to the Editor. Irish Medical Times, 11
February, p. 4.
Whelan, C. T., Hannan, D. F. & Creighton, S. (1991). Unemployment,
Poverty and Psychological Distress. Dublin: The Economic and Social
Research Institute.
Wing, J. K. (1990). The functions of asylum. British Journal of Psychiatry,
157, 822-7.
13
Lessons from America: semantics and services
for mentally ill homeless individuals
LEONA L. BAGHRACH
2 31
232
L. L. BAGHRACH
233
234
L. L. BACHRAGH
235
236
L. L. BAGHRAGH
Geographic mobility
There is now evidence that some severely mentally ill homeless people
are highly mobile: they simply move about a great deal (Lamb, 1982;
Appleby & Desai, 1987; Bachrach, 1987, 1992; Mercier et aL, 1992;
Dixon et aL, 1993; Wolch, et aL, 1993). Although some professional
writers appear reluctant to acknowledge the full extent of mobility
within this population (Bachrach, 1987), a number of popular
journalists have reported on the prevalence of this circumstance and
captured its significance (Bachrach, 1990).
In fact, three broad varieties of mobility are found among mentally
ill homeless persons. First, there is substantial movement into and out
of the homeless population, so that some individuals may be regarded
as 'permanently' homeless and others as 'episodically' homeless (Arce
et aL, 1983).
Secondly, there is considerable diurnal and seasonal movement of
the population within circumscribed geographical areas. Although
some mentally ill homeless people live more or less continuously in
one place, many others branch out within the community as services
become available to them or as their specific needs for food, shelter
and health care shift.
Thirdly, many mentally ill homeless people exhibit patterns of
gross mobility over wide geographical areas and move, with greater
or lesser regularity, from one city or region to another (Bachrach,
1987, 1988a).
The antecedents of mobility among these individuals are complex;
inadequately investigated, they are certainly appropriate areas for
intensive study. It appears, however, that some portion of the
restlessness that characterizes mentally ill homeless persons relates to
sociologically entrenched patterns of coping (Stokols & Shumaker,
1982). Some individuals in the population, for example, recall
237
238
L. L. BAGHRACH
239
240
L. L. BAGHRACH
241
Belkin, L. (1992). Treating the sick can mean clothing them too. New
York Times, 24 November, pp.Bl, B2.
Bragg, R. (1994). Homeless seeing less apathy, more anger. New York
Times, 25 February, pp. Al, B2.
Campbell, R. L. (1981). The language of psychiatry. Hospital and
Community Psychiatry, 32, 849-52.
242
L. L. BAGHRAGH
Kanter, A. S. (1989). Homeless but not helpless: legal issues in the care
of homeless people with mental illness. Journal of Social Issues, 45,
91-104.
Lamb, H. R. (1982). Young adult chronic patients: the new drifters.
Hospital and Community Psychiatry, 33, 465-8.
Lamb, H. R. (1990). Will we save the homeless mentally ill? American
Journal of Psychiatry, 147, 649-51.
243
14
Homelessness and mental health:
lessons from Australia
HELEN HERRMAN AND CECILY NEIL
Introduction
The extent of serious mental illness among single homeless people,
broadly defined, is similar in Australia to that found elsewhere in the
Western world (Scott, 1993). During the 1980s in Australia the
problems of ill and disaffiliated people became increasingly apparent
to service agencies. The problems have been less visible than in other
parts of the world, hidden to a large degree in special accommodation
houses, cheap boarding houses and shelters or crisis accommodation
centres (Doutney et aL, 1985; Herrman et al., 1989; Teesson &
Buhrich, 1990, 1993; Virgona et al., 1993). Even so, national reports
estimate that thousands of Australians, and in particular many
thousands of young people, live without adequate shelter (HREOC,
1989; Neil et al., 1992).
A relatively high proportion of homeless people have a mental
illness, but many do not. The problem of homelessness, for those with
or without severe mental illness, is influenced by the social and
housing issues that affect all marginalized people. At the same time it
is important to recognize among the homeless those who would
benefit from psychiatric treatment, and the barriers to care that exist
(Bachrach, 1987). Furthermore, homelessness among mentally ill
people may be seen in part as a reflection on inadequacies in the
delivery of health, welfare and housing services to those with severe
mental disorder, many of whom develop multiple disabilities and
handicaps, and to their families and other caregivers.
Homelessness was defined recently in an Australian government
report as: 'that state in which people have no access to safe and secure
shelter of a standard that does not damage their health, threaten their
244
245
246
247
suggestion that 'it is the articulate, those with higher incomes and
those with less demanding behaviour who have the best chance of
accessing government programs, whilst others of equal need are
forced into substandard private provision' (Victorian Council of
Social Service and the Office of the Public Advocate, 1992, p. 13).
Local economic conditions and social welfare policies may be seen
in this way to create a changeable 'threshold5 for true or absolute
homelessness among isolated people with severe mental illness. This
view is supported by recent work in the USA challenging the
impression that homeless mentally ill people are shelterless by choice.
Two prospective follow-up studies reported good clinical and social
outcomes for individuals provided with accommodation and on-site
mental health care and day care (Lipton et aL, 1988; Caton, 1990);
and assertive case management helped women who wished to move
around to remain in contact with services and be adequately housed
(Harris & Bachrach, 1990). The majority of 208 long-stay patients
discharged from mental hospitals in Sydney and elsewhere in New
South Wales to supported community accommodation between 1984
and 1987 were functioning well and had remained in the accommodation over some years (Andrews et ai, 1990).
In this chapter we will consider factors that may change this
threshold, as well as those that may affect the entry to and exit from a
broadly defined homeless state. Social and housing policies, and the
ways in which services for people with severe mental illness are
organized, are all likely to be relevant. Individuals are differently
affected according to many aspects of their own background,
personality and situation. The needs of people soon after the onset of
an illness and with a close and perhaps affluent family for support, are
different from those of a destitute person alone after several years of
episodically worsening disorder. The management and prevention
strategies possible will depend on the attitude and behaviour of those
working in and planning the services, the resources provided and the
social policy climate.
248
determined by a combination of broad social trends, family circumstances and individual attributes (Susser et al., 1988). Post-war
changes in homeless populations are commonly associated with
broad social changes. These include changes in the population
structure, a growing scarcity of housing, increased levels of unemployment and changes in welfare and psychiatric services. The
process of'deinstitutionalization' of the mentally ill has affected each
generation differently over the past 30 years. In Australia, as in other
countries, this process has been implemented overall without adequate treatment or support or residential alternatives, and appears to
have interacted with the political economy to contribute to homelessness among the mentally ill. Policies have taken little account
furthermore, of the needs of families, and of the factors that may
enhance the ability and willingness of families to provide care. The
breakdown of family links tends to increase the chances of homelessness for an individual with severe mental illness.
While wider factors may be involved, attention to risk factors at the
individual level may reverse, or prevent, a drift towards homelessness.
Homeless mentally ill people frequently have a history of childhood
family separation or institutional care (Susser & Struening, 1987),
and are often socially isolated (Thornicroft & Breakey, 1991). A
Melbourne survey of homeless and disaffiliated people found high
rates of overlap in individual respondents between substance dependence and the mood and psychotic disorders (Herrman et al., 1989).
Similar rates of dual diagnosis were found in other recent surveys
using standardized diagnostic instruments (Koegel et al., 1988).
Co-morbidity of mental disorders and of mental and physical
disorders may well be an important factor in the genesis of homelessness, as may cognitive impairment. Over 40% of a cohort of homeless
men resident in a large shelter in Sydney showed at least mild
cognitive impairment and more than one in four showed severe
impairment (Teesson & Buhrich, 1993).
In the relationship between risk factors and homelessness, the
direction of causality may be unclear. For instance, drug abuse, social
isolation and mental disorder are all plausible consequences as well as
causes of homelessness. In addition, it is usually difficult to date the
onset of homelessness for any individual (Susser et al., 1988). In the
Melbourne study, life charts were constructed from the information
obtained at interview, which was often rich although retrospective,
and from state records of service use. For many respondents evidence
of severe mental disorder preceded sustained periods of homelessness
249
250
20% and 40% of the persons surveyed in these studies have had a past
history of psychiatric hospitalization (compared with 40% of respondents in Melbourne) and another 15% had contact with outpatient
psychiatric services only. The higher rates of hospitalization have
applied in those studies that consider admissions for drug or alcohol
abuse. Very few of the people with a history of hospital stay have had
251
252
253
254
255
256
social services, specialist services frequently ignore communitymanaged housing groups and generalist support workers who may
have contact with their clients, other than to criticize their perceived
reluctance to provide accommodation for people with psychiatric or
drug-related disabilities.
The problems of co-ordination are exaggerated by the fact that
community-managed housing organizations may have to find their
way through a maze of possible services to help some of their clients
access the type of assistance they need. Further, the growth of the
community management of services has led to a degree of conflict
between specialist, professionally-oriented, community-based government agencies and community-managed organizations that
supply generic support. The former are concerned that professional
sensitivities are being ignored, and the latter fear that clinical care
will dominate the life of their clients.
There is now a greater recognition of the need for a holistic
approach on the part of specialist government agencies, as well as on
the part of community-managed housing groups, and therefore of the
need for more intersectoral linkages. However, there is still a general
lack of intersectoral co-ordination, and, indeed, of mutual respect,
common language and level of understanding between generic
service providers and providers of specialist services (Sawyer et al.,
1992).
257
258
259
An increase in the number of single bedroom units in mediumterm supported accommodation projects so other tenants already
under stress do not have to accept responsibility in crisis situations
for someone whose behaviour is disruptive or threatening.
More projects that provide accommodation specifically for people
with challenging behaviour, and have limited expectations of the
residents (McDonald, 1993).
260
Conclusion
Australia has a system of comprehensive health insurance and state
funded mental health services, which are currently being integrated
and linked into local systems of health and social services. Recognition of the plight of homeless mentally ill people has been part of the
impetus for a national approach to reform of mental health services.
However, economic recession and policies of economic rationalism
make the necessary reforms difficult to implement and in themselves
are likely to result in more mentally ill people becoming homeless and
shelterless. The genuine attempts to respond to need and to emphasize preventive approaches continue. Areas of particular interest in
this regard include the promotion and support of primary medical
practitioners in the care of people with serious mental illness, and
recognition of the needs of families and other caregivers.
Vital to service developments at this point are recognition of the
role of community psychiatrists and other key clinicians and support
workers in local service developments, and negotiation of the role and
resourcing of the various housing support sectors. Service research
and evaluation are required. Services need to be responsive to
community needs, and at the same time public education is required
about the scale of the problem, and about the service and policy
changes that are likely to help. Professional advocacy, and support for
self-help and family support groups, are important in the face of the
continuing stigma of mental illness.
References
Andrews, G., Teesson, M., Stewart, G. & Hoult, J. (1990). Follow-up of
community placement of the chronic mentally ill in New South
Wales. Hospital and Community Psychiatry, 41, 1848.
261
262
1085-92.
Lamb, H. R. & Lamb, D. M. (1990). Factors contributing to
homelessness among the chronically severely mentally ill. Hospital
and Community Psychiatry, 4 1 , 301-5.
263
264
186-91.
PART IV
POLICY AND EVALUATION
15
Implications of social policy
DAVID KINGDON
Housing
People with mental illness will be particularly vulnerable to being
made homeless. In a competitive market for homes, those with severe
and enduring mental illness will be disadvantaged by their lack of
employment, disabilities consequent on their illness and stigmatization. Securing accommodation may therefore be difficult and
maintaining it equally problematic for much the same reasons. In
turn, housing difficulties are life events and circumstances predisposing to relapse and persisting disability. Of particular consequence
therefore, has been the availability of low-cost housing in the public
or private rental sector. Both areas are known to have reduced over
267
268
D. KINGDON
269
Emergency shelters
A response to homelessness over many decades has been the development of emergency shelters. In New York City, Bassuk & Lauriat
(1986) reported there to have been 2 shelters before 1980 and 18 in
1986 and they asked the question: 'Are emergency shelters the
solution?' Their answer was that: 'Emergency shelters are an essential
short-term solution to the plight of the homeless. They save lives. The
trouble is that many shelters do little more ... Shelters [are] becoming permanent institutions that have replaced the almshouses and
mental institutions of past decades' (Bassuk & Lauriat, 1986, p. 134).
Some shelters in the USA have developed 'wards' staffed by psychi-
270
D. KINGDON
271
number of recommendations highlighting the need for local assessment of the service needs of mentally disordered offenders to take
account of the need for local and medium secure hospital provision at
all levels and non-secure provision. This includes efficient development of court diversion schemes and transfer of mentally disordered
offenders, where appropriate, to community support or hospitals and
the general development of links between the criminal justice system
and health and social services. Both appear to be occurring and in
1992/3 transfers to hospital increased substantially.
272
D. KINGDON
those who are mentally ill in particular. However, some leverage can
be applied through policy and management channels to improve
services and, crucially, develop understanding by PHCTs of the
difficulties faced by mentally ill people when they are about to, or
have become, homeless. Access to specialist care for most people is by
referral from the PHCT and this means that initial assessment,
treatment and emergency care is given prior to contact being
established with specialist services. The PHCT will usually have
records and often acquaintance with the individual that can be a
source of valuable information for specialists. Even more importantly, anxieties about the process of referral to mental health services can
be alleviated and any complications or disruptions in this process can
be resolved. The PHCT in effect acts as advocate and provides an
introduction for the person to services that elicit apprehension in most
people. For those who need it most, people with severe mental illness
who are also homeless, this smoothing of the path into specialist care is
usually not available. Many homeless people are not registered with a
PHCT. Where they are, this is often only temporary, previous records
are unavailable and time or circumstances frequently makes the
development of a positive therapeutic relationship difficult. Contact
with specialist services may therefore occur in difficult circumstances,
through Accident and Emergency services or the criminal justice
system.
In the UK, two pilot projects were established in 1986 in the City
and east London and Camden and Islington areas to improve
primary health care available to homeless people. Multi-disciplinary
teams were set up, each with a salaried general practitioner (GP),
who visited places where homeless people congregated, seeking where
possible to secure admission of homeless people to GPs lists. Evaluation of the scheme in 1989 suggested some drawbacks to this separate
approach to the homeless and modifications have been made to the
scheme, which was expanded in 1991, to integrate it more with
mainstream services.
273
274
D. KINGDON
275
276
D. KINGDON
277
Conclusions
When homeless people have been asked about their needs, food,
shelter and general medical care usually come before mental health
care, if the latter is included at all. Policies need to address these
expressed needs. But there also needs to be provision of appropriate
mental health care designed to prevent people from becoming
homeless and mentally ill and to assist them in maintaining their
independence once other basic needs are being met.
There is a longstanding tradition of local responsibility for care for
the homeless. Such responsibility can lead to services that are tailored
to local need because of more individualized assessment of needs. In
most areas, the local mental health teams are in the best position to
provide the range of services needed by the relatively small numbers
of homeless people with mental health problems in consultation with
them. Policies need to be focused on creating services that recognize
their needs with staff that have the basic level of training needed to
address them. This may involve management specification through
purchasing contracts to ensure targeting of services. In a few inner
city areas specialist workers or teams are needed to reach patients assertively, not aggressively - and reintroduce them to mainstream
services where possible. A few will need continuing support from the
specialist teams or workers. Likewise specialist hostels are needed in
some areas, direct access hostels are essential in the short-term but
cannot be expected to provide longer term mental health care.
Interest in homelessness and mental illness by policymakers has
dramatically increased in the past 5 years (Manderscheid & Rosenstein, 1992). Early signs are that this may be having an impact on
numbers and availability of services. Evaluative studies are underway
and will report in the near future.
Acknowledgements
Acknowledgements are due to colleagues from the various different
government departments involved for their comments on early drafts
of the above.
278
D. KINGDON
References
Access to Health (1992). Purchasing and Poverty. London: Access to
Health.
Bachrach, L. (1992). What we know about homelessness among
mentally ill persons: an analytical review and commentary. Hospital
and Community Psychiatry, 43, 453-64.
Barham, P. (1992). Closing the Asylum. Harmondsworth: Penguin.
Bassuk, E. L. & Lauriat, A. (1986). Are emergency shelters the solution?
International Journal of Mental Health, 14, 12536.
279
16
Evaluating services for homeless people with
mental disorders: theoretical and practical
issues
M A X MARSHALL
Introduction
This chapter is in two sections. The first section will be a survey of the
types of evaluative studies that have been conducted on services for
homeless people with mental disorders. The survey will pay particular attention to the problems that have arisen in carrying out these
studies. This section will be illustrated throughout by examples of
evaluative studies from the UK literature; where UK studies are
lacking, examples will be taken from the world literature.
The second section will consider how far evaluative studies have
provided evidence for the effectiveness of hostels for the homeless. As
indicated in Chapter 9 the role of hostels in this area is increasingly
controversial.
Retrospective evaluations.
Evaluations based on the impressions of a trained observer.
Surveys.
Follow-up evaluations.
Before and after evaluations.
Single case and 'action research' evaluations.
Quasi-experimental evaluations.
Randomized controlled trials.
Evaluating services
281
Retrospective evaluations
Description
Retrospective evaluations are based on the analysis of routine data
collected during clinical work.
Uses of retrospective
evaluations
Examples
An example of a retrospective evaluation is the description by
Ferguson & Dixon (1992) of a walk-in psychiatric service in a hostel
for the homeless in Nottingham. An example of a retrospective
evaluation that raised questions about the effectiveness of a service, is
the study by Hamid & McCarthy (1989) of community psychiatric
nurses (CPN) working with 'homeless' and 'home-based5 clients in
Bloomsbury. This study found that homeless clients were less likely to
receive supportive care from the CPN service and were more likely
than home-based clients to be referred elsewhere; even though the
homeless group were more likely to suffer from schizophrenia.
Advantages
Retrospective evaluations are relatively cheap and fairly easy to carry
out.
Disadvantages
Retrospective evaluations are usually based on poor quality information and are prone to many sources of bias. Whilst retrospective
282
M. MARSHALL
Examples
A good example of the use of an observer is provided by a study of the
reaction of agencies to 'difficult to place' homeless clients in Oxford,
most of whom were suffering from mental disorders (Vagg, 1992).
The observer concluded that 'many of the most difficult problems
were picked up by the voluntary sector, which might be characterized
as the only place left to go when statutory agencies decided that
individuals had become too difficult to cope with'. As a result of the
observer's recommendations a care management team was set up to
organize and supervise 'shared care' in the city, between voluntary
and statutory agencies.
Evaluating services
283
Advantages
Evaluations based on impressions are relatively cheap, though not as
cheap as retrospective studies. Perceptive observers may provide
information that cannot be obtained from more formal empirical
approaches. For example, observers can detect good or bad practices,
or gain an understanding of the motivations or frustrations of staff
and clients. The findings of observers may lead to hypotheses that can
then be tested in empirical studies. Such findings may also be used to
complement or 'explain' the findings of empirical studies. For
example, a randomized control trial might establish that a new
service was no more effective than usual treatment, whilst the findings
of an observer might help explain why this was so.
Disadvantages
The study of primary care teams in London (Williams & Allen, 1989)
illustrates well the problems of evaluating a new service by means of
an observer. Without suitable empirical data to back up the impressions of the observer it is impossible to counter the argument that the
observer's impressions were misleading or biased. The inevitable
result of a negative report will be a wrangle between proponents and
opponents of the service. Thus, observational methods, whilst of value
in identifying problems with existing services, are not helpful when
used as the sole means of assessing the effectiveness of a new service.
This is not to say however that they should not be used to complement
a suitable empirical approach, such as a randomized control trial.
284
M. MARSHALL
Surveys
Description
Surveys are based on data collected from a cross-section of the users of
a service. The subjects selected should be representative of the users of
the service (but rarely are).
Examples
Eliciting clients3 opinions
Evaluating services
285
Advantages
Surveys are useful in that they provide a good means of assessing the
quality of care being provided by a service at one point in time.
Surveys also permit us to evaluate a service indirectly by providing
information about how the recipients of the service perceive it.
Disadvantages
Surveys can be difficult to conduct, particularly when representative
samples are required and standardized instruments are used. Surveys
do not allow us to assess the actual outcome for clients but only the
quality of care at one point in time. Surveys cannot determine
whether one approach is superior to another. Where quality of care is
being assessed, it may be very difficult to find a suitable yardstick
against which the care of the index group may be compared.
Follow-up evaluations
Description
A group (or cohort) of clients receiving a service is identified and
assessed at one point in time and is then reassessed some time later.
Uses offollow-up
evaluations
286
M. MARSHALL
Examples
A Bristol follow-up study reported a favourable outcome in 16 out of
48 homeless men who had attended a psychiatric clinic (Tomison &
Cook, 1987). The study concluded that 'psychiatric intervention can
achieve something'. On the other hand, the East London Homelessness Health Team (HHELP), a primary health care team, reported
that only 8 out of 112 subjects of no fixed abode, and 12 out of 144
subjects in unstable accommodation, obtained stable accommodation during their period of contact with the team (Balazs, 1993).
Advantages
Follow-up evaluations are easier to conduct than randomized controlled trials and quasi-experimental designs. Such studies complement surveys because they assess the outcome of care rather than the
process of care. Follow-up evaluations provide a reasonable indication of how far a service is achieving its goals, and may indicate when
a service is performing poorly.
Disadvantages
Follow-up evaluations are difficult to carry out with homeless persons
because the high mobility in this group leads to high drop out rates.
Furthermore good outcome for clients cannot necessarily be attributed to the effects of the service, whilst a poor outcome may reflect
factors outside the control of the service in question. Nevertheless such
studies give some indication of how well homeless people are being
served by the current system of care.
Evaluating services
287
Examples
The Psychiatric Shelter Program of the Presbyterian Hospital in New
York City was evaluated by comparing the social circumstances of 32
homeless mentally ill men before and after the men were placed by the
programme in community housing (Caton et aL, 1990). Findings
were a reduced level of criminal justice contacts and an increased
utilization of psychiatric after care services.
Advantages
Before and after evaluations are a useful method of making a
preliminary assessment of a new service when either: (a) more
suitable designs are not possible, or (b) the investigator wishes to
determine whether it is worth proceeding to a controlled trial. The
simple design of before and after studies makes them relatively easy to
conduct.
Disadvantages
There are likely to be high drop out rates in samples taken from
homeless populations. Furthermore, before and after evaluations
tend to over-estimate the effectiveness of an intervention because,
without suitable control groups, it is impossible to rule out the
possibility that changes in the index group were not due to natural
improvements over time (regression towards the mean), or to other
factors that changed during the study. This second confounding
factor can be a particular problem in studies of the homeless because
of the frequent policy changes that occur towards this group.
288
M. MARSHALL
Evaluating services
289
Quasi-experimental evaluations
Description
In quasi-experimental evaluations the outcome for an index group of
subjects (who receive a service) is compared with that for a control
group of subjects (who do not receive the service). The allocation of
subjects to index and control groups is not random.
Uses of quasi-experimental
evaluations
Examples
No examples from the literature on homelessness are known to this
author, although such designs have been used in studies of homebased care (Dean et aL, 1993).
Advantages
Quasi-experimental evaluations avoid the need for randomization,
with its attendant ethical and administrative problems.
Disadvantages
Failure to randomize subjects to treatment and control groups means
that differences in outcome between groups may be attributed to
factors other than the intervention. Quasi-experimental evaluations
290
M. MARSHALL
are prone to bias and their findings are therefore much less robust
than the findings of randomized controlled trials.
Evaluating services
291
Subjects were referred from: hostels for the homeless; night shelters; a
GP clinic for the homeless; the City Council homelessness unit; and
local voluntary sector group homes. Eighty subjects consented to be
randomized to treatment or control groups. At 14 month follow-up
there were no significant differences between treatment and control
groups in number of needs, quality of life, employment status, quality
of accommodation, social behaviour and severity of psychiatric
symptoms. The authors concluded that social services case management was successful in reducing deviant behaviour, but otherwise
made little difference to the lives of the subjects.
Advantages
The randomized controlled trial is the most robust way of producing
definitive evidence for the effectiveness of a service or for its
superiority over another service.
Disadvantages
Randomized controlled trials are difficult to organize under the most
favourable conditions. Unfortunately evaluations of services for the
homeless tend to take place under conditions that are particularly
unfavourable to organizing a successful clinical trial. The greatest
problem is caused by high drop-out rates, either because subjects
have moved away from the area altogether, or have changed
'address' locally and therefore lost contact with the investigators.
Even where an index intervention leads to low drop-out rates, a study
may fail because of high drop-out rates in the control treatment, as
was the case in the first trial described above. The second trial
reported above, attempted to overcome the problem of high drop-out
rates by replacing subjects who dropped out with other subjects
randomly allocated to the three treatment groups (Morse et aL, 1992).
Further problems likely to face those attempting to organize
randomized controlled trials with homeless mentally disordered
subjects are: insufficient numbers of subjects; difficulties obtaining
consent to randomization; and difficulties obtaining co-operation for
randomization from voluntary organizations.
292
M. MARSHALL
Hostels
Hostels
Hostels
Hostels
Evaluating services
293
(7000 in London alone in the past 10 years) has greatly impaired the
capacity of hostels to continue performing this vital role. This has led
some to conclude, not unreasonably, that 'the increasing numbers of
homeless people on the streets have been generated primarily by the
disappearance of direct access accommodation' (Timms, 1993).
The success of hostels as a safety net appears to be for three reasons.
First, large to medium hostels usually have beds immediately
available. Secondly, they do not impose complex referral or assessment criteria before admission and thirdly, they are usually easily
located by potential users.
Providing a place to live
Whilst conditions in some hostels are unacceptable, they are not
invariably poor. Nevertheless the available evidence suggests that
most mentally disordered hostel residents would prefer to live
elsewhere. For example, in a study of women's' hostels in London 16
out of 61 women (26%) expressed a preference to remain in the
hostel; moreover all 16 who wished to remain complained about lack
of privacy (Marshall & Reed, 1992). The author's own work in
Oxford hostels (unpublished results) indicates that less than one in
three male residents with severe mental disorder would prefer to
remain in the hostel. We must assume therefore that many hostel
residents would prefer to live elsewhere.
Providing a resettlement service
There is considerable evidence to suggest that hostels are not
successful at resettling residents with severe mental disorder in more
suitable accommodation. For example, a follow-up of mentally
disordered hostel residents in Oxford found that only 10 out of 48
residents were rehoused in an 18 month period. Those residents who
were rehoused went either: back to their families, to private bedsits, or
to accommodation provided by the hostels. No residents obtained
accommodation supported by health, social services or housing
associations (Marshall & Gath, 1992). In this respect there has been
little improvement from the days of the Camberwell Reception
Centre (Wood, 1976). It would appear however that this failure to
resettle residents is not so much a failure of the hostels as a failure of
the statutory services to support the hostels.
294
M. MARSHALL
Evaluating services
295
CONCLUSION
The available evidence suggests that hostels for the homeless are
successful in one major aspect of their role; that of acting as a safety
net for mentally disordered people who would otherwise have
nowhere to go. However, on the other three aspects of the hostels' role
the evidence is not so favourable. On providing a place to live, the
limited evidence suggests that only a minority of mentally disordered
residents wish to remain in hostels, and few find that the accommodation provided is satisfactory. On acting as a resettlement service, the
evidence suggests that hostels are largely unsuccessful. On acting as a
place where care and support can be provided, the available evidence
suggests that hostels are not succeeding.
The limited effectiveness of hostels is not, however, a strong
argument for their immediate closure. By collecting and concentrating people with mental disorders in one place, hostels have provided,
and continue to provide, a useful opportunity for statutory agencies
to fulfil their obligations. The signal failure of statutory agencies to
take up this challenge is hardly the fault of the hostels. What is
required is a much more dynamic and aggressive effort by statutory
agencies to identify, treat and especially resettle, mentally disordered
hostel residents. Until this effort materializes and hostel beds are lying
empty, we should leave the messenger unharmed.
References
Balazs, J. (1993). Health care for the single homeless. In K. Fisher (ed.),
Homelessness, Health Care and Welfare Provision, pp. 51-93. London:
Routledge.
Caton, C. L. M., Wyatt, R. J., Grunberg, J. & Felix, A. (1990). An
evaluation of a mental health program for homeless men. American
Journal of Psychiatry, 147, 286-9.
Dean, C , Phillips, J., Gadd, E. M., Joseph, M. & England, S. (1993).
Comparison of community based service with hospital based service
for people with acute, severe, mental illness. British Medical Journal,
307, 473-7.
Ferguson, B. & Dixon, R. (1992). Psychiatric clinics in homeless hostels
- your flexible friend. Psychiatric Bulletin, 16, 683-4.
Hamid, W. A. & McCarthy, M. (1989). Community psychiatric care for
homeless people in inner London. Health Trends, 21, 67-9.
Hogg, L. I. & Marshall, M. (1992). Can we measure need in the
296
M. MARSHALL
homeless mentally ill? Using the MRC Needs for Care Assessment
in hostels for the homeless. Psychological Medicine', 22, 1027-34.
Leach, J. & Wing, J. K. (1978). The effectiveness of a service for
helping destitute men. British Journal of Psychiatry, 113, 481-92.
Marshall, E. J. & Reed, J. L. (1992). Psychiatric morbidity in homeless
women. British Journal of Psychiatry, 160, 761-9.
Marshall, M. (1989). Collected and neglected: are Oxford hostels for the
homeless filling up with disabled psychiatric patients? British Medical
Journal, 299, 706-9.
Marshall, M. & Gath, D. G. (1992). What happens to homeless
mentally ill people? Follow up of residents of Oxford hostels for the
homeless. British Medical Journal, 304, 79-80.
17
Future directions for homeless mentally ill
DlNESH BHUGRA
The lessons from a literature review from around the world suggest
that the definitions of homelessness vary but the association between
homelessness and chronic severe mental illness is fairly well known.
The health services also vary according to the local socio-economic
and political structures, which means that the true assessment of
actual numbers of the homeless and their needs presents a major
problem. In the UK, localized catchment area based community
services may enable a closer primary and secondary care interface as
described in Chapter 11. In the USA, marked reduction in low-cost
housing and deinstitutionalization of state and county mental hospitals have contributed to the problem of social construction of
homelessness (Robertson & Greenblatt, 1992). The global view, not
comprehensive by any means, outlined in this volume suggests that
there are common themes for research, management and social policy
change and that there are also differences in terms of definitions of
poverty and homelessness, which may be inextricably linked, and also
in terms of service provision. If any strategies for change in research,
social policy and management are to succeed there have to be
concentrated efforts on the part of the researchers, clinicians, policy
makers and politicians. As the causes of homelessness are multidimensional, their management as well as social policy changes must
be multi-factorial and multi-dimensional.
298
D. BHUGRA
Definitions
Many policy makers seem to think that the homeless are a homogeneous group. Many sociologists and clinicians along with epidemiologists have struggled with the diversity of causes of homelessness, types
of homelessness complicating the presence of mental illness. There
needs to be an element of agreement in definitions of homelessness as
well as classification. Leach (1979) had urged a simple intrinsic versus
extrinsic divide - the former including those where the mental or
physical disability antedated their homelessness, whereas in the latter
their homelessness was attributed to situational factors. Leach went
on to propose that the subgroups would need different service
programmes. However, Morse et al. (1991) suggested that a
taxanomic system based on current impairment was superior to the
system based on past history. Obviously the roofless versus shelterless
definitions will vary across countries. Furthermore, the presence of
co-morbidity will affect the definitions and the outcomes. There need
to be multi-centred studies across the globe addressing the issues of
definitions, pathways and antedating factors.
Future directions
299
300
D. BHUGRA
Definitions of poverty
In the developed countries, where social services exist, the official
poverty line may determine the social services support and benefits.
In less developed countries, these basic supportive networks may not
be present and a combination of poverty, overcrowding and industrialization may lead to child labour as well as to large scale
migrations into urban areas producing large areas of shanty towns.
Here, houses may have nothing more than a basic structure with
walls and a roof, but no availability of basic amenities for clean water
or sewage disposal. Furthermore, the truth behind the definitions
may be used by the policy makers to hide the extreme poverty or
exaggerate it to fulfil certain aims. The men, women and children
who must live on the streets and shelters may find little privacy and
less support. The homeless are only the most visible of the poor. The
much greater number of very poor who cling to vermin-infested
rooms or live doubled and tripled up with relatives are otherwise little
different in the USA (Hilfiker, 1989). Although Hilfiker (1989) goes
on to observe that in the USA the poor and marginalized must fend
for themselves - this is true of most of the globe - certainly from
countries where information is openly available. A lack of resources
and denial of access to these resources compound the problems of
poverty and its impact on people of different ages, ethnicity and
gender. These factors need to be studied in relationship to homelessness.
homelessness
Future directions
301
factors and these would need to be assessed along with the other
vulnerability factors. Lower educational attainments and lack of
financial savings have been suggested to be related to inability to
postpone gratification (see Bhugra, 1993 for a review). The estimates
of psychiatric morbidity in the homeless vary. Co-morbidity of
physical illness, psychiatric illness and substance misuse although
well described does not always translate into an understandable
structural model. In developing countries, for example, physical
illness and chronic infections may lead to poverty, unemployment or
homelessness. However, it is fair to say that the evidence of a
correlation between homelessness and alcoholism or mental illness
and homelessness does not mean that either factor is a cause or
consequence of homelessness (Wright, 1989). In the UK disorganized
schizophrenia is common in homeless populations, it is not clear
whether this is true elsewhere as well. The social substrate of health
perceived by the homeless individual will also affect the relationship
with the therapist.
Patterns of help-seeking
Patterns of help-seeking depend upon the type of illness, the explanatory model used for that particular illness, the types of service
available and accessibility to those services. It would thus make sense
that homeless mentally ill across the globe may have different
patterns of help-seeking. This, when combined with the perception of
social substrate for health, affects help-seeking. The individual may
display a limited trust in statutory services. Furthermore, in the UK,
sectorization and emphasis on catchment area services along with a
perceived rigidity and inflexibility of the health services, especially
secondary care, mean that the homeless mentally ill may not always
be booked into health services. In addition, the theoretical model of
money following the patient after the National Health Service (NHS)
reforms means that any definition of last home may prove difficult
and unacceptable, thereby stopping the 'follow the patient flow' of
the money. If the individuals are not registered with general
practitioners, a hand-over and subsequent continuity of care will
prove impossible. Some homeless individuals blame their nonregistration on their own mobility and the inflexibility of services.
The innovative outreach services can therefore prove to be more
accessible, approachable and therefore successful. In other countries,
302
D. BHUGRA
Structures
The physical and political structures of the health and social services
provide help for the homeless mentally ill individual. These structures
of the health care systems also include personal/social and complementary health sectors, which may be used by those needing health
care. In the USA it was estimated that between 70% and 90% of all
illness episodes are treated within the personal/social sectors, which
may be a reflection of the general health care system available. In
other countries, including the UK, such information is not readily
available, but another possible area of investigation across various
centres may well be the extent of use of alternative therapies and other
options. Nevertheless, not having any fixed, easily available, approachable social support, the homeless individual may have to rely
more on personal resources. This may also produce extra stress if these
Future directions
303
Future directions
There remain several avenues in research that can be followed. There
are two areas that could immediately be researched fruitfully - these
include age-related differences and special needs groups within the
broad mentally ill homeless category. In addition, some further
suggestions are made below.
Impact of urbanization
In less developed countries the impact of urbanization on numbers of
homeless mentally ill individuals as well as social and psychological
stressors and functioning needs to be studied. Although initial
welcome steps have been taken in Latin America, the Far East,
Middle East and South East Asia still need to be studied (for a
discussion on Latin America see Harpham & Blue, 1995). As
Harpham (1993) had previously cautioned, risk factors such as
poverty and poor environmental conditions have been repeatedly
identified as having an independent association with ill-health among
the urban poor in general, but their impact across various cultures
and in various settings is not always very clear. Maternal education,
maternal age, parity and marital status for the women have been
shown to be key features associated with ill-health, and have already
been demonstrated by Harpham in Brazil. Another additional factor
that needs to be assessed is that of intra-urban differences. Social drift
and social residue have been shown to be two possible factors for the
intra-urban differences (Ekblad, 1990). The pull towards the bright
lights of the city and push factors from the rural poverty need to be
studied. It will be worth investigating which of the two factors plays a
more important and sustained role in the genesis and perpetuation of
mental disorders.
304
D. BHUGRA
individual vary a great deal but there is a lot that can be adapted by
other countries. For the young homeless for example, the model
demonstrated by Van der Ploeg (1989) in the Netherlands suggests
various dimensions of negative family backgrounds, past history of
professional help along with negative school experiences, low selfperception and lack of friends, although reports from elsewhere also
suggest that there may be generalities that are common across
countries and cultures and there are specifics that must be taken into
account. Similarly the models presented in this volume suggest that
not only is there an accumulation of severely mentally ill in the
lodging houses, there is also a possibility of reduction in numbers due
to availability and government commitment to providing low-cost
accommodation to those in need, as demonstrated by reports from
Denmark. Similar numbers in Eire suggest various social factors at
play.
Anthropological approaches
Koegel (1992) argues that the dearth of rich qualitative descriptions
of lives in the processes of homelessness, together with an emphasis on
preoccupation with pathology and disaffiliation, a failure to view the
homeless in the broader socio-economic and situational contexts of
their lives and the absence of longitudinal perspectives make it more
difficult to provide relevant social policy information. He goes on to
observe that the danger in using epidemiological techniques when
focusing on homeless individuals is that such individuals are isolated
from the broader socio-economic contexts that are crucial in shaping
the characteristics, behaviours and choices. Thus, the explanatory
models highlighting critical influences of broader factors, settings and
other people rather than an individual deficit model can give an
insight into the processes of homelessness. Furthermore, studying
individuals over a period of time in an ethnographical context means
that processes and insights into their experiences of double jeopardy
of mental illness and homelessness can work. In addition, the
anthropological method of participant observation moves beyond an
exclusive reliance on self-report by working on the principle that
behaviour is being studied using a blend of both interviewing and
observation (Koegel, 1992).
Future directions
305
Summary
For service provision, the gap between the needs of the homeless
mentally ill and the available resources and services have to be
matched. The multi-dimensional problem needs multi-dimensional
solutions and any medical interventions planned will have to be
highly tailored. Furthermore, housing, welfare benefits and service
delivery have to be interlinked. Joint planning between the statutory
health services and voluntary services may produce unique innovative models of assessment and management. As Craig et al. (1995)
recommend, the needs of the newer subgroups within the homeless
population (younger, ethnic minority and women) should be targeted. There is no doubt that preventative work, especially if it is
multi-disciplinary, will prove to be fruitful.
References
Bhugra, D. (1993). Unemployment, poverty and homelessness. In D.
Bhugra & J. Leff (eds.), Principles of Social Psychiatry, pp. 355-84.
Oxford: Blackwell Scientific Publications.
Burns, A. & Bhugra, D. (1995). History and structure of the National
Health Services. In D. Bhugra & A. Burns (eds.), Management for
Psychiatrists, pp. 3-17. London: Gaskell.
Burrows, L. & Walentowicz, P. (1992). Homes Cost Less Than
Homelessness. London: Shelter.
Craig, T., Bayliss, E., Klein, O., Manning, P. & Reader, L. (1995). The
Homeless Mentally III Initiative. London: DOH.
Ekblad, S. (1990). Family stress and mental health during rapid
urbanisation. In E. Nordberg & D. Finer (eds.), Society,
Environment and Health in Low-income Countries. Stockholm: Karolinska
Institute.
Harpham, T. (1993). Urbanisation and mental disorder. In D. Bhugra
& J. Leff (eds.), Principles of Social Psychiatry, pp. 346-54. Oxford:
Blackwell Scientific Publications.
Harpham, T. & Blue, I. (1995). Urbanisation and Mental Health in
Developing Countries. Aldershot: Avebury.
Hilfiker, D. (1989). Are we comfortable with homelessness? Journal of the
American Medical Association, 262, 1375-6.
Koegel, P. (1992). Through a different lens: an anthropological
perceptive on the homeless mentally ill. Culture, Medicine and
Psychiatry, 16, 1-22.
Leach, J. (1979). Providing for the destitute. In J. K. Wing & R. Olsen
306
D. BHUGRA
Index
Abram Man 15
law enforcements 49
ACCESS (Access to Community
low self-perception 53
Care and Effective Services
mental health 52-3
and Supports) 125
negative experiences at school 53
accommodation
negative family backgrounds 53
cheap rented, diminished stocks
runaways
and homelessness 34
definitions 48-9
and psychiatric service, lack of
throwaways and homeless 48
co-ordination 255-6
services to avoid homelessness
supported, and service provision,
253
changes required for
see also young homeless
co-ordination 258-60
advisory services for homeless,
see also hostels; housing; residential
Germany 204
planning; shelters
affective disorders
address, mailing and GP
amongst older homeless, London
registration 162
and NYC 156
admission unit, outcome, Ireland
in hostel users 22
217-20
African-American homeless women
adolescents, homeless 48-53,113
67
between social security and social
African origin and homelessness 36
department care 49
age and mental illness 249
common causes for running
age factors, homeless women 645,
away 49-53
65-6
differentiation between street
see also older homeless
youth and homeless youth
Aid to the Disabled 269
52-3
AIDS and homeless adolescents 51
few friends 53
alcohol abuse/alcoholism 112
history of professional help 53
amongst homeless, Denmark
implications for service 53-4,
190, 192, 193
55
by mentally ill homeless 104
institutional care 49
and criminality 82
307
308
alcohol abuse/alcoholism (cont.)
homeless women 69-71
19th-20th centuries 63-4
and homelessness 35
in hostel users 22
older homeless 155
and other drug dependence in
homeless 11415
parental 155
and young homeless,
Denmark 194
as pathway to homelessness 172
in single homeless 171
statistics 30
see also dual diagnosis
aliases used by patients 217
ambulatory difficulties, older
homeless 156-7
America see United States of
America
amnesic syndrome amongst mentally
ill hostel residents 146
anthropological approaches 304
Assertive Community Treatment
(ACT) 126
Assesment Service, Ireland 21625
asylums for homeless, Denmark
190-1
asylums, management of 'lunatic
paupers' 16-18
extra cost for maintaining 17
attitudes of mentally ill homeless
101-2
to professional services 113,
116-17
attitudes to mentally ill homeless by
mental health professionals
4, 19-22, 102-3
Australia, homelessness, and mental
health 244-64
Baltimore Mental Health Systems,
Inc. 126
barbiturates taken by homeless
women 70-1
Index
'Bedlam' 15
beds (bed spaces)
availability, and discharge of the
mentally ill 33
lack of long-term facilities 33
in direct access hostels 19
begging, prosecutions,
1910-1930s 80-1
behaviour disorders, homeless
children 47
Bethlehem hospital, fate 15
Bethlem asylum 16
bisexuality and homelessness 36
Black Death, cause of social
disruption 12
black men, drug abuse, and
homelessness 35, 36
black people, older homeless 154
black women
drug abuse, and homelessness
35, 36
homeless 66-7
board-and-care homes 273
boarding-homes for homeless,
Denmark 190-1
boarding-houses, Germany 205
Boards of Guardians 14
boundary-busting system, USA
238, 240
bridewells (houses of correction)
12-13
first workhouses 13
Britain, homeless women, 19th and
20th centuries, 59-63
broken homes
and family homelessness 44
and homeless adolescents 50
Camberwell Reception Centre 293
accommodation studies 137
Camberwell Spike, closure, effects
19
cannabis abuse and homelessness
35
see also 'soft' drugs
Index
Care Programme Approach 275
care system for homeless
Germany 204-5
and homeless youth 50
see also community care; primary
health care; psychiatric
services
Caribbean origin and homelessness
36
casual wards
closure, Ireland 211-12
provision 13
reduction by National Assistance
Board 14
renamed reception centres see
reception centres
for single homeless women 62
casualty departments for primary
care of single homeless 179
Cecil Houses (Inc.) Public Lodging
Houses Fund 62
cellulitis, reason for presenting at
medical services 173
Child Care Act 1991, Ireland 213
Children Act of 1908, Ireland 213
children, homeless 458
behaviour disorders 47
classification 11213
developmental delay 46, 47
emotional problems 46
Ireland 213
language development delay
46-7
medical disorders 47-8
missing school 48
poor school achievement 46
statistics 456
suicidal thoughts 46
see also young homeless
children of older homeless, contacts
with 157-8
Christ House, programme 126-7
Church Army
homeless women accommodation
62-3
309
hostel accommodation 134
Church leaders and homelessness,
Ireland 216
classification
adolescents 113
runaways, throwaways 48-9,
113
chronic alcoholics 112
chronically mentally ill 112
dually diagnosed 112
of homeless children 112-13
of homeless families 112-13
of homeless women 645
of homelessness 28-38, 172
in 1920s 20
'intrinsics' and 'extrinsics' 28
see also homelessness, definitions
situationally homeless 112
street people 112
clinical services for homeless people,
USA 110-32
acceptance of the way of living
120-1
basic needs 117-18
priorities do not always include
mental health services
117-18
clinical considerations 120-1
clinical and support range 118
engagement 116
establishing trust 120
future plans 127-8
prevention 128
giving help and training to
'providers' 121
integration 119-20
in localized areas 11617
peculiar problems for people
living in streets 119
programmes, models and service
systems 122-7
dual diagnosis 123-4
Federal programmes for the
homeless mentally ill 124-5
increasing accessibility 122
310
clinical services (cont.)
innovative programmes at the
local level 125-7
linkage and collaboration
122-3
reluctance to accept change
121
rights of individuals 118-19
team work 121
transition stage 119
see also primary health care;
psychiatric services
Clinton, President Bill, health care
plans 274
co-morbidity
Germany 204
in the homeless 115-16
see also dual diagnosis; physical
health
co-ordination of services 102,113,
118, 121, 180-1, 255-6, 275,
276
cocaine abuse
by homeless women 116
by mentally ill homeless 104
and homelessness 35
cognitive deficits (impairment)
amongst older homeless, London
and NYC 156
in homeless mentally ill 103
Columbia University in New York,
programme 126
Commissions of Inquiry on Mental
Handicap and Mental Illness,
Ireland 215
Committee on Reformatory and
Industrial Schools, Ireland
213
community care
dangerousness of discharged
patients 92
Germany 201-5
local level, services 125-7
police referrals to psychiatric
services 84-6, 271
Index
community mental health
programmes 1045
community-oriented approach for
integrated psychiatric
services 2568
community psychiatric nurse
service 281
community psychiatric nurses, UK,
statistics 274
community psychiatrists, UK,
statistics 274
community treatment order and
subsequent criminality
89-91
compulsory treatment 90-1
extended leave patients 90-1
'concealed homeless' 62
County Homes, Ireland 211
crack abuse and homelessness 35
criminal activity and homeless
adolescents 51-2
criminal justice system and
legislation, impact on
mentally ill homeless 270-1
criminality, homelessness, and
mental illness 7895
access to primary care 889
alcohol problems 82
amongst older people 155
community treatment order
89-91
diversion from custody 86-8
future provisions 92
historical background 79
psychiatric hospitalization 82,
82-3
discharge problems 83
public concern at dangerousness
of discharged patients 92
studies 81-3
see also police referrals
criminality, single homeless 173
Crisis at Christmas emergency
shelter, morbidity studies
171-2
Index
Critical Time Intervention (CTI)
126
custody, diversion from for mentally
ill defendants 86-8
Cyrenian shelter, Oxford 136
dangerousness of discharged
patients 92
dangerousness of street living 114
day centre, St Brendan's Hospital,
Dublin 220-5
community based components
224
day care and accommodation
needs 224-5
operational aspects 220-2
outcome 222-3
outcome data 223
day centre and shelter for homeless
women 74
definition (meaning) of homelessness
see classification, of
homelessness; homelessness,
definitions
deinstitutionalization
as contribution to homelessness
4, 31, 34, 160, 211, 267-8
Germany 199-200, 201
and homelessness 100
and inadequate housing and care,
Australia 246, 248
mentally ill need appropriate
housing 246-7
user-pay principles 2467
numbers affected 2734
and older homeless, London and
NYC 161
and type of service offered 235-6
in urban settings, Ireland
215-16
dementia
amongst homeless, Denmark 192
amongst mentally ill hostel
residents 146
amongst older homeless, London
311
and NYC 156
demographic profile of single
homeless 172
Denmark, homelessness 18997
characterization 1913
homeless mentally ill 1913
definitions 189-90
institutions for the homeless
190-1
psychiatry in the streets 1945
research in mental illness among
homeless 1956
statistics 190, 191
use of psychiatric central register
192-3
young homeless 193-4
Department of Social Security
(DSS) resettlement units 13
depression, amongst older homeless,
London and NYC 156
deserving/non-deserving and
attitudes to homelessness 4
destitution and workhouse
accommodation 13
detention of homeless mentally ill
270-1
developmental delay, homeless
children 46, 47
disruptive behaviour, effects on
other residents 259
domestic conflict and homeless
adolescents 49
drop-in centres
Germany 205
for older homeless 165
see also day centre
drug abuse
amongst homeless, Denmark 193
homeless women 69-71
age factors 70-1
and homelessness 35
and mental illness 248
in single homeless 171
statistics 30
see also dual diagnosis
312
Index
introduction 187-9
see also Denmark; Germany;
Ireland; United States of
America
evaluation of services see services,
evaluating
'extended leave' patients 90-1
Index
funding at local level 13
funding policies 259
funding shortages and treatment
254-6
future directions for homeless
mentally ill 297-305
future prospects, Australia 256-60
313
lack of data 198
Nazi massacre of mentally ill
199
statistics 201-3
studies 202, 203-4
Goddard Riverside Community
Center 120
government programmes for the
homeless, USA 124-5
see also statutory health services
Great Chapel Street Medical
Centre, Soho, London
continuity of care 177
open access to all services 178
primary care health centre for the
homeless 88
guilt at circumstances 174
314
homeless mentally ill, and mentally
ill homeless, differentiation
99-100
homeless people
acceptance of way of life 120-1
dangerousness of street living
114
other problems (mailing address,
storage of medicine . . .) 119
programme, Ireland 21620
admission unit 217-20
day centre 220-5
see also Day Centre, St
Brendan's Hospital, Dublin
prospective survey, Ireland 218-19
UK, statistics 135
see also hostels; shelters
homeless population in the USA
112-16
see also classification
homeless, single see single homeless
homeless women see families with
children, homeless; women,
homeless
homelessness
classification 28-38
see also classification, of
homelessness
definitions 26-8, 48-9, 151-2,
230-3, 298
Australian government 2445
commonalities across the
globe 297-9
first, second and third degree
245
how soon is a person homeless
27
increased by shortage of funds
and treatment 2546
introduction to the problem 3-9
historical aspects 4-6, 11-25
international aspects 3-4
overview and definition of
clinical, social and
psychological needs 7-9
Index
range of psychiatric morbidity
5-6
meaning 230-3
and mental health, Australia
244-64
and mental illness, evaluating
services see services,
evaluating
models 26-40
National Assistance Act 1948
requirements 26, 268
pathways, see pathways into
homelessness
as psychiatric rather than political
problem 215-16
statistics 152
see also adolescents, homeless;
children, homeless; families
with children, homeless
homosexuality
and homeless adolescents
51-2
and homelessness 36
hospitalization
enforced, rights of individuals
111-12
of older homeless 157
hostels
British, survey of mental illness
and alcoholism 21-2, 22
clients' opinions 284, 293
closure 'programme' 18-19
definition of shelter and hostel
accommodation 1334
direct access
bed spaces 19
UK 270
resettlement units 270
effectiveness 292-4
estimating numbers 143-5
evaluating 280-96
see also services, evaluating
follow-up studies 141
mixed shelter/hostel studies
137-8
315
Index
numbers and characteristics of
residents 143-7
older homeless 158
rehousing 141
residents
psychiatric characteristics
146-7
316
institutions for the homeless,
Denmark 190-1
integration of services for single
homeless 1801
international aspects 3-4
see also Denmark; European
perspectives; Germany;
Ireland; United States of
America
intravenous drug abusers, and
homelessness 35
involuntary help for mentally ill,
NYC 270
involuntary treatment (admission)
118-19
involuntary treatment for mentally
ill homeless 111-12
Ireland, homelessness 209-29
admission unit, outcome
217-20
closure of casual wards 211-12
day centre 220-5
operational aspects 220-2
deinstitutionalization in urban
settings 215-16
disinvestment in public housing
212
domestic perspectives 210
emigration 211
homeless children 213
homeless families 214
homeless females 214
identifying the homeless 210
lack of accommodation 211
outcome of the Irish initiative
216-17
'planning for the future' 225-6
poverty, morbidity and the
'underclass' 212
prison service 214
social factors 210-11
statistics 210
survey 216
tuberculosis incidence 223-4
Irish people
Index
background of mentally ill hostel
residents 146
origin and homelessness 36
preponderence amongst single
homeless 173
proportion of homeless, London
154
itinerant workers 20
language development delay,
homeless children 467
leg ulceration
reason for presenting at medical
services 173-74
treatment difficulties 178-9
legislation and the criminal justice
system, impact on mentally
ill homeless 270-1
life-time disorders 249
local authority
duty to house the homeless 13,
268
availability of funds 13
local community level services
125-7
local economic conditions and
homelessness 247
see also statutory health services
lodging houses
for homeless, Denmark 190-1
mixed common 60-1
for respectable working women
60
women considered more difficult
than men 601
London, older homeless, see older
homeless, London and NYC
low-cost housing see housing, low-cost
'lunatic paupers' 15-18
Bethlehem hospital 15
curable, management 17-18
management of the furiously mad
and dangerous 16-17
treatment in asylums 17-18
treatment in workhouses 17-18
317
Index
'Vagrancy' Act 1714
'wandering' 15
15-16
318
methadone abuse, and
homelessness 35
Metropolitan Poor Act 1964 13
missing school, homeless children
48
mobility, geographic, of mentally ill
homeless 236-8, 250
diurnal and seasonal movement
236
into and out of homeless
population 236
over wide areas 236
reasons for restlessness 236-7
whilst undergoing treatment 237
mobility problems, older homeless
156-7
models of homelessness 26-40
monasteries, relief to the poor 12
dissolution by Henry VIII 12
mood swings and homeless
adolescents 52
morbidity and homelessness,
Ireland 212
MRC Needs for Care Schedule 284
multiplicity of needs 113
see also co-ordination of services
National Assistance Act 1948 26-7,
268
National Assistance Board
and Casual Wards 14
establishment 14
men/women living in reception
centres, 1966 62-3
networks, lack for homeless people
113
neuroses, amongst homeless,
Denmark 193
New York, older homeless, see older
homeless, London and NYC
New York City Homeless
Emergency Liaison Project
(Project HELP) 270
night-lodgers, Ireland 211-12
no fixed abode
Index
psychiatric admissions 69
alcoholism 71
women 69
statistics for personality disorders
21-2
statistics for schizophrenia 20-2
non-itinerant workers 20
Northern English preponderence
amongst single homeless 173
numbers of homeless mentally ill,
world-wide 299
see also statistics
occupations, older homeless, London
and NYC 154
older homeless, London and NYC
150-69
cross-national comparison
151-68
access to benefits 161-2
access to health care 162
case advocacy 164
causal model of ageing and
homelessness 163
demographics 153-4
deviant behaviour 155
disruptive experiences in
youth 154-5
drop-in centres 165
emergency shelters 165
family contacts 1601
low cost housing 160
mental health 155-6
methods 152-3
patterns and pathways of
homelessness 158-9
physical health 156-7
preference for safe
accommodation 165-6
psychiatric hospitals, release of
patients 161
social networks 157-8
solutions 163-6
statutory entitlements 162
uptake 162
319
Index
statutory support 163-5
structural (social) factors
159-62
theoretical approach 151
unemployment 161
definitions 151
numbers 150-1
organic mental disorders
amongst mentally ill hostel
residents 146
amongst older homeless, London
and NYC 156
Orwell, George 17-18
parish
relief for poor 79
returning vagrants to 79
PATH (Program for Assistance in
Transition for the
Homeless) 124
pathways into homelessness 29,
30-8, 298-9
contribution of mental illness 31
definition 29, 30-8
deinstitutionalization 4, 31, 34,
160, 211, 267-8
economic factors 33-4
epidemiological difficulties 32
families with children 41-5
individual factors 347
older homeless 158-9
severe mental illness 36-7, 247-9
statistics 29-30
unemployment 33
women 723
personality disorders
amongst homeless, Denmark 192
in the homeless 115, 116
in single homeless 171
statistics 21-2
physical health
homeless children 47-8
in homeless women 71-2
of older homeless, London and
NYC 156-7
Ireland 225-6
police involvement with mentally ill
people 271
police referrals to psychiatric
services 84-6, 271
characteristics of referrals 85-6
subsequent remand factors 87-8
policies, development of
co-ordinated services
required 275
policy changes and psychiatric
service provision 256-60
Poor Law Act 1930 14
Poor Law administration 14
Poor Law Amendment Act of 1834
13
Poor Law Commissioners report of
1842 16
Poor Laws, and mentally disordered
homeless, historical factors
80
post-traumatic stress disorder in the
homeless 116
poverty
definitions 300
and family homelessness 42
and homeless families 43
and homelessness 233
Ireland 212
level, calculation 159
and income benefits 159
as pathway to homelessness 33
relationship of patterns to
homelessness and mental
illness 300-1
predisposition to homelessness
154-5
preference for safe accommodation
by older homeless 165-6
prevalence rates
schizophrenia 135, 139-41, 171
320
prevalence rates (cont.)
of specific psychiatric disorders
114
prevention of homelessness 128,
274-5
Australia 254-6
for mentally ill 250
primary health care
access by homeless mentally ill
88-9
amongst older homeless 157
medical centres 889
for the mentally ill homeless
251-2, 271-2
for older homeless 162
services, use by homeless mentally
ill 251-2
primary health care of the single
homeless 170-83
aspects of the consultation
174-7
building trust 175-6
continuity of care 176-7
demographic and psycho-social
profile 173
doctor's communication with
patient 175-6
engagement with the patient
174-5
factors underlying presentation to
medical services 173
failure of projects 179-80
humanity of carers 175-6
morbidity profile 170-2
need for integration 180-1
presenting problems, new
patients 171
registration, GPs attitudes 157,
177
role of the receptionist 175
various types of provision of
health services 177-9
see also general practitioners
primary health care teams
(PHCT) 271-2
Index
priorities of homeless people 101,
113, 117-18
prison background
amongst mentally ill hostel
residents 145
homeless women 67
homelessness and mental illness
82
Ireland 214
of reception centre residents 812
single homeless 173
prison population increase since
community care policy 845
prisoners, mentally disordered
attempts to rescue 86
conditions endured 86
Home Office guidelines on
remand 86
remand delays and psychiatric
assessment 86-7
private sector accommodation for
homeless women 62-3
Program for Assistance in Transition
for the Homeless (PATH)
124
Programme for the Homeless,
Ireland 216-25
admission unit, outcome 21720
day centre 220-5
see also day centre, St Brendan's
Hospital, Dublin
programmes for the homeless
122-7
government 1245
see also services
Project HELP 270
Project Outreach 120
prostitutes, women,
accommodation 61
prostitution and homeless
adolescents 501
provision and use of mental health
services, Australia 24953
psychiatric admissions, emergency
85-6
321
Index
psychiatric assessment, Germany
204
psychiatric assessment service
London courts, results 87
Magistrates' Court 87-8
psychiatric attitudes to mentally ill
homeless 19-22, 102-3
psychiatric central register,
Denmark 192
psychiatric characteristics of hostel
residents 1467
psychiatric hospitals see
deinstitutionalization; mental
hospitals
psychiatric illness see mental illness;
pathways into homelessness
psychiatric morbidity levels and
homelessness 31, 32-3
psychiatric problems amongst single
homeless 172-3
psychiatric rather than political
problem 215-16
psychiatric services
and accommodation services, lack
of co-ordination 2556
failure to assess broad range of
need 102
insufficient, Germany 1989
provision, changing policies
associated with 256-60
see also clinical services
psychiatry and the homeless
mentally ill, historical
aspects 19-22
psychiatry in the streets, Denmark
194-5
psychopathia, amongst homeless,
Denmark 193
psychoses
among prison population 193
amongst Danish lodging house
residents 193
treatment difficulties 178-9
psycho-social profile of single
homeless 172
322
residential planning (cont.)
idea of social isolation important
239-40
mobility of mentally ill homeless
236-8
special funding 239
special needs 2356
type of service offered 235-6
see also hostels; shelters
respiratory disease in single
homeless 170
restlessness and mentally ill
homeless 236-8
'revolving door' group 86
rights of individuals 11112,
118-19,270
risk factors and mental illness 248
Rowton Hotels 14
Royal College of Psychiatrists
Working Party Report 1991,
categories of homelessness
28-9
runaways 48-9, 113
safe accommodation, older
homeless 165-6
St Brendan's Hospital, Dublin,
survey of 425 admissions
216-20
see also Ireland, homelessness
St Mungo's Community in London
evaluation 288
hostel accommodation 135,
140
St Peter's workhouse, Bristol 16
Salvation Army
description of stay 61-2
homeless women 60
statistics 62-3
hostel accommodation 134, 136
new admissions 137
hostel closure, effects 19
work with poor and homeless
13-14
schizophrenia 113
Index
among the homeless, 1900-1960s
19-22
in hostel population 22
amongst mentally ill hostel
residents 146
amongst older homeless 155
amongst young homeless,
Denmark 194
female preponderance 345
high prevalence in single
homeless 171
in homeless women 68-9
in hostels
apparent rise in level 141-3
reclassified studies 141-3
estimating numbers of people
143, 144, 145
higher prevalence 139-41
possible explanations 140-1
as pathway to homelessness 31,
172
prevalence
hostel accommodation 139-41
UK studies 135
with substance abuse 113,
123-4
schizophreniformic psychosis,
amongst homeless,
Denmark 192
school achievement, poor, homeless
children 46
Scottish background of mentally ill
hostel residents 146
Scottish origin and homelessness
36
Scottish preponderence amongst
single homeless 173
security against violence 147, 165
self-discharge from hospital 217
self-harm amongst mentally ill hostel
residents 147
semantics and services for mentally
ill homeless, USA 230-43
separated parents, and
homelessness 35-6
Index
services 99-109
clinical, for the homeless, USA
110-32
see also clinical services for
homeless people, USA
Denmark 189-97
see also Denmark, homelessness
engagement and outreach 105
evaluating 280-96
advantages and disadvantages
281-2, 283, 285, 286, 287,
288, 289, 291
based on impressions of a
trained observer 282-3
before and after evaluations
286-7
description 281, 282, 284-5,
286, 287-8, 289, 290
effectiveness of hostels 292-4
see also hostels
eliciting clients' opinions 284
examples 281, 282-3, 286,
287, 288, 289, 290-1
follow-up evaluations 285-6
quasi-experimental
evaluations 289-90
randomized controlled trials
290-1
retrospective 281-2
single case and 'action research'
evaluations 287-9
survey of studies 280-91
surveys 2845
uses 281, 282, 284, 285, 287,
288, 289, 290
factors limiting access 252-3
Germany 197-209
see also Germany, homelessness
and homeless mentally ill 1001
for homeless women 73-4
USA studies 73-4
lack of trust in statutory services
100, 101
long-term intensive case
management 106
323
mental health and rehabilitation
services 1067
need for unstructured flexible
approach 102-3, 105,
106-7
need for wide range of housing
options 107
provision and use, Australia
249-53
conflicting views of providers
253
culturally-sensitive facilities
253
factors limiting access 252-3
fragmentation 253
'harm minimization' 253
provision where needed 106
psychiatric see psychiatric services
reasons for non-attendance
101-2, 101-2
and semantics for mentally ill
homeless, USA 230-43
for single homeless 177-81
failure 179-80
need for integration 180-1
see also clinical services; primary
health care; psychiatric
services
specialized, Melbourne 251
specific problems 101, 1024
attitudes 102-3
models of mental health care
delivery 102
substance misuse and dual
diagnosis 103-4
voluntary services 104
use 101-2
what is required? 1047
sexual assaults and homeless
adolescents 51
sexual orientation and
homelessness 36
sexually offensive behaviour
amongst mentally ill hostel
residents 147
324
Index
shelters 133-48
and day centres for homeless
women 74
definition of shelter and hostel
accommodation 1334
emergency, for older homeless
165
Germany 205
mixed shelter/hostel studies
137-8
325
Index
young homeless 41
statutory health services
entitlements, older homeless,
London and NYC 162
uptake 162
see also income benefits; local
authority
and lack of help for mentally ill
homeless 301-2
and non-statutory sectors,
co-ordination 276
structures 302-3
support for older homeless,
London and NYC 163-5
street living
not preferred by older homeless
165-6
as only definition of homelessness
151
substance abuse 103-4, 113
amongst homeless, Denmark 193
and homeless adolescents 50
and homelessness 35
need for services by mentally ill
homeless 100
see also dual diagnosis
suicidal thoughts, homeless
children 46
suicide attempts and homeless
adolescents 52
Supplemental Security Income 269
Supplementary Benefits Act 1976
270
see also income benefits
surveys see services, evaluating
Sydney, special problems 2456
Task Force on Homelessness and
Severe Mental Illness 124,
275
temporary accommodation 4
The Psychiatric Services - Planning for
15-16
326
Vagrancy Act 1824 79-80
amendment of sleeping out
offence 80
vagrants, returned to 'place of
settlement' 80
Victorian era, management of
mentally ill homeless 1119
violence amongst mentally ill hostel
residents 147
voluntary agencies, definition of
homelessness 26-7
voluntary services 104
whipping for lunatics forbidden 15
white men, drug abuse, and
homelessness 35
white/non-white, older homeless,
NYC 154
white women, homeless 66-7
women with children and
homelessness 43-4
see also families with children
women, homeless 59-77
19th and 20th centuries
Britain 59-63
after Industrial Revolution
59-60
casual wards (renamed
reception centres) 623
interwar period 61
living-in as condition of
employment 60
lodging houses for respectable
working women 60
mixed common lodging
houses 60-1
temporarily stranded/
permanently homeless 61
Victorian era 60
United States of America
63-4
differences in problems 63-4
Skid Row 63-4
statistics 63
age factors 645, 65, 65-6, 66
Index
alcohol problems 69-71
black/white women 66-7
causes 72-3
characteristics 65-7
Denmark 191-2
with dependent children 66
drug problems 6971
higher level of serious psychiatric
morbidity 345
Ireland 214
marital status and homelessness
35
mental illness 67-9
London, non-statutory project
67-8
prevalence studies 67
schizophrenia 68-9
physical illness 71-2
prison/institution background 67
proportion 153
services 73-4
USA studies 73-4
types 645
younger than homeless men
65-6
workhouses 4
evolution 1115
beginning of segregation of
sexes and conditions 1314
dread by working classes 13
general mixed 13
houses of religion 12
monasteries 12
punitive attitudes 13-14
expenditure and grants for
insane 17
and homelessness and criminality,
19th century 79
management of 'lunatic paupers'
16-18
not considered good for the mad
17-18
Working Party on Single
Homelessness in London
(SHIL) 292
Index
Wytham Hall, admission of patients
with complex problems 179
Wytham Hall residential care home
for the homeless 89
young homeless
Denmark 193-4
327
statistics 41
see also adolescents, homeless;
children, homeless; families
with children, homeless
youth, disruptive experiences and
homelessness when older
154-5