Homelessness and mental health

populations in a continuum of unsatisfactory housing from cardboard boxes through long-stay hostels. Not surprisingly, therefore, estimates of the size of the homeless population vary enor mously from survey to survey depending on the definition and on less scientific influences of a political or campaigning nature. For example, estimates of the numbers of homeless people in England and Wales range from around 2000 at any point in time when denned as people sleeping rough to as many as 75 000 if extended to those in hostels, squats and bed and breakfast (Office of Population Censuses and Surveys, 1991; Moore et al. 1995). Similarly, estimates of the numbers of hospital admissions for people of no fixed abode vary widely, partly because of differences in the use of the term between psychiatrists even when they all work within one health authority (Cowan & MacMillan, 1996). While these issues of definition call for great care in interpreting data, there is a consensus that, however defined, there has been an explo sion in the numbers of homeless people in Britain's major cities during the past decade.


Tom Craig
The term 'homeless' has been used to describe populations in a continuum of unsatisfactory housing from cardboard boxes through long-stay hostels. Not surprisingly, therefore, estimates of the size of the homeless population vary enor mously from survey to survey depending on the definition and on less scientific influences of a political or campaigning nature. For example, estimates of the numbers of homeless people in England and Wales range from around 2000 at any point in time when denned as people sleeping rough to as many as 75 000 if extended to those in hostels, squats and bed and breakfast (Office of Population Censuses and Surveys, 1991;Moore et al. 1995). Similarly, estimates of the numbers of hospital admissions for people of no fixed abode vary widely, partly because of differences in the use of the term between psychiatrists even when they all work within one health authority (Cowan & MacMillan, 1996).
While these issues of definition call for great care in interpreting data, there is a consensus that, however defined, there has been an explo sion in the numbers of homeless people in Britain's major cities during the past decade. The number of households placed in temporary accommodation by local authorities doubled from 23 000 in 1986 to over 40 000 in 1989, and the numbers claiming board and lodging payments quadrupled between 1979and 1986(Central Statistical Office. 1991. This expansion was largely accounted for by young men, women, families and ethnic minorities which paralleled wider changes in the social economy -a scarcity of low-cost housing, high unemployment, the erosion of traditional family networks and down sizing in the organisation and delivery of supportive services (Bachrach, 1984;Craig & Timms, 1992). Of all these factors, the shortage of entry-level accommodation is probably the most important. There was an 85% decline in the production of new council housing output between the mid-1970s and the end of the 1980s that coupled with a shortfall in predicted voluntary and private sector supply amounted to a net loss in London alone of 243 000 rental this issue. units between 1981 and 1988: a figure that continued to rise right through the early 1990s (Greve, 1991). In comparison to their domiciled peers, young homeless people are not only less likely to obtain independent housing but are also less likely to have successfully completed basic education, less likely to have ever held employ ment and far more likely to have experienced parental neglect, abuse and rejection throughout their childhood, with as many as 40% having been in children's homes and other institutions (Fischer et al, 1986;Cohen & Thompson, 1992: Craig et al, 1996. While there is therefore reasonable evidence to support the impression of a rise in the numbers of homeless people and by extrapolation, an increase in the numbers with a mental illness, the changing demography of the population is also associated with a rather different profile of psychiatric disorders to that described in earlier studies of homeless populations, with affective disorders, substance dependence and personal ity disorder being the most frequently reported problems. Comorbid mental illness and sub stance use disorder is the rule rather than the exception and this creates major problems for treatment and rehabilitation (Shaffer & CatÃ³n, 1984;Craig et al, 1996).
It has also been suggested that the increase in the number of mentally ill homeless people occurred as a result of the closure of the psychiatric hospitals and failures in the com munity care alternatives that were designed to take its place. But there is good evidence that this was not a very important factor, fewer than 5% of long-stay residents lost to follow-up following the closure of Friern Barnet were thought to have had a spell of homelessness (Leff, 1991); and only 1 in 10 homeless people with a functional psychosis seen in the Londonwide Homeless Mentally 111Initiative had spent more than 12 months in hospital in their lifetime, the most common experience being of multiple, brief admissions (Craig et al, 1995).
But if the closure of the psychiatric hospitals did not directly contribute to the problem, there can be little doubt of the importance of deinstitutionalisation in the wider sense. The community reprovision of the old psychiatric hospital bed catered solely for the existing long-stay patients with no immediate capacity to cope with newly accumulating cases of patients whose disorders have failed to respond to treatment and who are unable to manage independently. There are relatively few community-based facilities that can cope with the level of disability and disruption of many patients with chronic, treatment-resistant psy choses and even fewer 'wet' hostels that will tolerate comorbid mental illness and substance use disorder. In many instances, bed and breakfast is the only viable residential option for these most vulnerable cases, from which it is all too easy to drift out of contact with mental health services. The result has been the many well-documented failures in the care of patients discharged from acute wards and in the treatment histories of many severely ill homeless people. Not only has there been a failure to provide enough suitable residential alternatives to hospital accommodation but what little already existed may well have been inadvertently lost. In the decade preceding 1991, there was a net loss of 75% of direct access hostel spaces. These old direct access hostels housed many chronically ill people, largely out of contact with psychiatric services. Unlike the residents of the old psychiatric hospitals, these patients were moved on with out any recognition of their resettlement needs in terms of psychiatric treatment or specialised support (Craig & Timms, 1992).
Despite these gloomy observations, there is evidence that services can be arranged and provided in ways that make a difference. The Rough Sleepers Initiative, launched by the Department of the Environment in 1990 and aimed at boosting resettlement services to young people sleeping rough in central London, has been credited with a substantial success, the increased provision of temporary and permanent accommodation under this initiative being matched by a steady fall in the numbers of rough sleepers since the initiative was launched (Ran dall & Brown, 1996). At the same time, the Department of Health and the Mental Health Foundation jointly funded the Homeless Men tally 111Initiative, which succeeded in bringing specialist services to severely mentally ill home less people who had fallen through the net of community care with modest social and clinical benefit (Craig et al, 1995). The initiative spawned a number of other projects nationally with broadly comparable results (e.g. Commander et al 1997).
Although these initiatives have been broadly successful, there are several aspects of clinical practice that remain to be resolved. First, it seems clear that lengthy and labour intensive efforts are required to engage the homeless person in treatment which often only proceeds once the basic necessities of food and shelter have been addressed. This flexible, non-coercive strategy for engagement must be balanced by the parallel duty to maintain a therapeutic focus, the lack of which is sometimes linked to libertarian tendencies to emphasise the patients' right to refuse treatment rather than his right to receive treatment. Second, while there is consensus on the need for a multi-disciplinary input, there is little consensus on the therapeutic options that should be available within teams and relatively few are able to offer equally skilled interventions for mental illness and substance dependency despite the common co-occurrence of these disorders in the target population. Third, generic teams, operating a single keyworker system may not be enough to ensure continued service uptake, probably because basing services around a single lead agency still excludes too many facets of care or is still too dependent on the severely disabled person having the personal resources to keep the string of appointments with separate service providers. A better model may involve specialist multi-disciplinary teams that provide all the necessary housing, social and medical care from one location and which are available for extended hours and at week ends. The best evidence for the efficacy of such an approach comes from North America where there have been several attempts to modify the assertive community treatment model of case management with some quite impressive results, including reductions in the length of time spent in homeless accommodation, improvements in psychiatric symptoms and in social functioning (Lipton et al, 1988: Morse et al, 1992.
While there is likely to be a continued need for specialist services for homeless mentally ill people for many years to come, the fact remains that homelessness among those with mental illness is a preventable adverse outcome that ought to be addressed by mainstream services. Proactive working with in-patients who are homeless or lose their accommodation during an admission, proactive work with landlords of patients whose tenure is precarious and the effective implementation of prison and court diversion schemes ought to be a good place to start.

Seminars in Alcohol and Drug Misuse
Edited by Jonathan Chick and Roch Cantwell Psychiatric complications of alcohol and drug misuse must be recognised and treated confidently. This book provides information to meet the needs of practising psychiatrists and will be useful to physicians, psychologists and social workers. A clear review of the aetiology, epidemiology, treatment and prevention of dependence on and misuse of alcohol and illicit and prescribed drugs is presented. With a balance of theory, recent research and practical clinical guidelines, the book covers specific and common problems in mental health as well as in general medicine. Â£13.50, 1994, 256pp