Who's sleeping in my bed?

The number of psychiatric beds in Friern Hospital in North London was steadily reduced in line with plans to close the hospital in 1993. Meanwhile some discharged patients deteriorate, most of whom have schizophrenia, with significant and multiple disabilities of a chronic nature and are ser vice dependent as described in the recent report of the Royal College of Psychiatrists (1993). Return to their accommodation may be prejudiced by their relapse and the attendant disturbed behaviour. The poor employment prospects too are unhelpful in rehabilitation, particularly since we are running down and closing probably one of the best indus trial rehabilitation facilities in the world. The com bined effects generate great pressures on the remaining beds and there are increasing numbers in prison. This study looks at the difficulty of discharging patients after recovery from an acute illness episode, when eligibility for accommodation must first be assessed by a social worker. A freeze on staff recruit ment, alongside social services budget reductions, has seen a progressive fall in the number of social workers in Haringey from an average of 174 in 1987 to 122 in 1988, and 77 in August 1989. These figures understate the problem as the cuts have been particu larly applied to the field social workers with whom clinicians must liaise for accommodation for their patients. The closure of a residential unit in the com munity and the reduction of day-care facilities have compounded the problem of inadequate resources, adding further discharge delays for many patients. The net result has been an increasing occupation of acute psychiatric beds because of accommodation difficulties. We look at the impact of these changes through data relating to patients admitted under one consult ant in West Haringey but believe that the findings can be generalised more widely in London.

The number of psychiatric beds in Friern Hospital in North London was steadily reduced in line with plans to close the hospital in 1993. Meanwhile some discharged patients deteriorate, most of whom have schizophrenia, with significant and multiple disabilities of a chronic nature and are ser vice dependent as described in the recent report of the Royal College of Psychiatrists (1993). Return to their accommodation may be prejudiced by their relapse and the attendant disturbed behaviour. The poor employment prospects too are unhelpful in rehabilitation, particularly since we are running down and closing probably one of the best indus trial rehabilitation facilities in the world. The com bined effects generate great pressures on the remaining beds and there are increasing numbers in prison.
This study looks at the difficulty of discharging patients after recovery from an acute illness episode, when eligibility for accommodation must first be assessed by a social worker. A freeze on staff recruit ment, alongside social services budget reductions, has seen a progressive fall in the number of social workers in Haringey from an average of 174 in 1987 to 122 in 1988, and 77 in August 1989. These figures understate the problem as the cuts have been particu larly applied to the field social workers with whom clinicians must liaise for accommodation for their patients. The closure of a residential unit in the com munity and the reduction of day-care facilities have compounded the problem of inadequate resources, adding further discharge delays for many patients. The net result has been an increasing occupation of acute psychiatric beds because of accommodation difficulties.
We look at the impact of these changes through data relating to patients admitted under one consult ant in West Haringey but believe that the findings can be generalised more widely in London.

The study
All patients admitted under one consultant (MPIW) in the Haringey area were included and discussed at the weekly ward round. The ward team formed a consensus view as to why each remained in the ward, assigning them to categories of mental illness, convalescence and accommodation difficulties. The accommodation category was used sparingly and excluded those with behavioural problems or medical or legal complications. The correspondence between up to eight independent observers on 13 patients was very satisfactory (Kappa 0.802).
The following dates were noted: week of admission, week of discharge, and week in which the patient would have been discharged but for lack of suitable accommodation.
All patients admitted prior to 1July 1987, includ ing four who remained in hospital throughout the study period, were excluded, as were those whose length of stay was less than three days.

Findings
The total numbers of beds available for the West Haringey district (population 120,000) include those in acute admission wards, rehabilitation, long-stay and psychogeriatric wards (Fig 1). Following reorganisation of beds in December 1987,there were, temporarily, 12 extra beds available. The loss of 31 beds in October 1988 represents the closure of a rehabilitation ward. There were 45 acute beds which remained static during the study period, with 79 to 100% bed occupancy, but consistently over 90% after October 1988.
During the two years there were 302 acute admissions of which 91 were readmissions. Of the 302 admissions, 286 were discharged before 1 July 1989, 15 to emergency bed and breakfast accommodation, three on more than one occasion.
The number of patients whose discharges were delayed each month is shown in Fig. 2, and the mean number of weeks by which their stay in hospital was prolonged for this reason in Fig. 3. Table I breaks this down according to diagnostic category and mean excess stay for each group. The number of beds occu pied by patients who are well is increasing (Fig. 4), indicating a possible displacement of the ill by the homeless in the nearly full-to-capacity acute admission beds.
Men were significantly more likely than women to be unmarried, have a psychotic illness, and to have a delayed discharge due to accommodation reasons. Unmarried patients were more likely to be psychotic and some tended to stay on the ward for accommodation purposes.

FIG. 3. Mean number of weeks by which discharge delayed
Psychotic patients spent more time on the ward for medical reasons and in total. Patients of non-UK origin spent more time on the ward for medical reasons, which reached significance for the males only.

Comment
The present policy at Friern, as at other hospitals facing closure, focuses on reprovision for "longstay" patients i.e. those continuously in hospital for longer than one year (reduced to six months follow ing these findings). There has been little involvement with those not coming under the "reprovision umbrella".
The increasing use of psychiatric beds as tempor ary accommodation is ironic in the light of the planned closure of the hospital with the transfer of long-stay patients' care to social services, as proposed in the Government White Paper (1989). The high cost of in-patient care at Friern (around Â£13,000per annum at the time of the closure pro posal), now close to Â£20,000with the reduction in patient numbers) was the main determinant of the closure, yet there remains a "perverse incentive", identified by the Audit Commission, to accommo date patients in hospital in order to alleviate pressure on local authority provisions. We find that some patients have been staying in hospital on average nearly seven weeks longer than required for medical purposes, while they wait for suitable accommo dation. These mean figures disguise wide variation and some patients can be eight or more months in hospital before even being assessed by a social worker.
Social services have a duty to provide suitable accommodation for the vulnerable homeless, includ ing those patients discussed here, after discharge. However, as stated in the Report of the Director for Social Services for Haringey in 1988, "... the budget cuts and consequent reduction in staff numbers, plus lowered morale and increasing staff sickness has resulted in social workers struggling under increased pressure, and Area Teams have been able to provide only an emergency service: Doors are often closed ... difficulties with organising opening hours as emergencies are unpredictable. One Area Team was unable to provide any duty service at all for several months ... the Council is not fulfilling its statutory obligations..." Cuts in social services obviously add to the diffi culty of patient care and also help create "new longstay" patients. Bed and breakfast type accommo dation is available for "emergency" purposes but is generally unsuitable for patients recently recovered from acute psychiatric symptoms and is often many miles from the hospital. For these reasons we have been reluctant to discharge patients as homeless and have sought instead to keep them in hospital until they have been assessed and more suitably placed, as recommended by the Minister of Health (Clarke, 1989).
Patients and medical staffare caught between the pressure to discharge due to bed closures and the decrease in social services provision. The decreased proportion in the use of acute psychiatric beds by acutely ill psychiatric patients (Fig. 4) indicates that it may be becoming easier to resist potential new admissions than to discharge patients lacking social care in the community. The situation is also causing pressure in other areas, such as the prison service (Coid 1988).