Originally published in BMJ 1993;306:669-670 [Medline]

The coining of the term "cardboard city" reflects the public consciousness of the visible increase in the number of homeless people on our cities' streets. Because of problems of ascertainment, putting an accurate figure to this total is impossible. Using its restrictive definition of homelessness, the Department of the Environment estimates that the number of homeless families in England has more than doubled over the past decade to 146,000 last year. (The housing charity Shelter says that this represents at least 420,000 people.)

Many of these people are likely to have serious psychiatric problems. An early study of the Camberwell Reception Centre (now closed) found that one third of the residents were suffering from a severe psychiatric disorder, excluding alcoholism.1 Twenty years later, in a survey of men arriving at a Salvation Army hostel a quarter were diagnosed as having schizophrenia.2 During the intervening period surveys of homeless people in residential settings have estimated the prevalence of severe mental illness at between 25% and 45%. While the proportion of mentally ill among homeless people has probably not changed greatly over the past two decades, the absolute number has probably doubled in parallel with the number of homeless people.

The number of occupied psychiatric beds in England has fallen from a peak of 148,000 in 19543 to about 45,000 in 1992. Some commentators have ascribed the rise in the number of homeless mentally ill people to the run down of psychiatric hospitals and have called for a halt to the policy of closure.4 But have the erstwhile long stay patients swelled the ranks of homeless people? There are two sources of evidence bearing on this question: one is the psychiatric history of homeless mentally ill people, while the other is follow up studies of long term psychiatric patients discharged into the community.

Only a few of the surveys of homeless mentally ill people have inquired about any history of psychiatric treatment. A study of people using a shelter in Boston found that although 30 of the 68 respondents were suffering from a psychosis, only five had been admitted to hospital for longer than a year in total.5 A survey of 124 homeless men using a Salvation Army hostel found that over half were suffering from mental disorders but only seven had been admitted to hospital for longer than a year.2 In an experimental study of a case management service for homeless mentally ill people only two out of 94 people referred in a year reported having been admitted to hospital for a year or longer.6 During the first year of the London homeless mentally ill initiative 544 homeless people were identified as suffering from severe mental illness. Of these, only three had had a continuous hospital admission lasting more than five years.7

The follow up studies of discharged patients have produced complementary evidence. The Team for the Assessment of Psychiatric Services has been evaluating the run down of Friern and Claybury Hospitals in north London. A follow up of the 278 long stay patients discharged between 1985 and 1988 failed to trace only six people, who were presumed to have become vagrants (three had been vagrants before their admission8). Of the 216 patients discharged in 1989-90, none became vagrants during their first year of follow up (J Leff et al, unpublished study). In a similar study of the closure of Cane Hill Hospital in south London, of 103 patients followed up one year after discharge, none had become homeless.9

These two sources of evidence suggest that the discharge of long stay patients from the declining mental hospitals is not the main factor contributing to the increase in the numbers of mentally ill people living on the streets. What other factors might be operating? One of the economic changes that is leading to increasing homelessness is the disappearance of low cost rented accommodation. People suffering from severe psychiatric illnesses are particularly vulnerable to this change because of their high level of unemployment, low earning power, and lack of social support.

Another possible contributory factor is the increasing difficulty in admitting acutely ill patients. The 1983 Mental Health Act, while safeguarding patients' civil rights, has made many mental health professionals reluctant to use its compulsory powers except in extreme cases. Furthermore, psychiatry has been affected by the general reduction in admission beds. Resulting deficiencies in the treatment of acutely ill patients can place intolerable burdens on their families and increase interpersonal friction, which may culminate in patients leaving home abruptly and being unable or unwilling to find alternative forms of shelter.

The fact that severely mentally ill people roam our streets without adequate accommodation and medical care is an inescapable indictment of our society. Instead of assuming that the discharge of patients from psychiatric institutions is to blame we should be studying carefully the pathways that lead mentally ill people into homelessness and destitution. If we are successful in identifying these the opportunities may exist to prevent a situation that is desperately difficult to cure.