The Mental Health National Service Framework (Department of Health, 1999) proposes in a headline target that individuals needing admission to hospital should be treated as close to home as possible. This would allow family and community links to be sustained and improve the prospects for integration of the hospital and community phases of care (Reference Glover and BindmanGlover & Bindman, 2001). One could reasonably assume that the same principle should also apply to any placements in specialist community residential homes.

Owing to an absence of appropriate residential services, people with learning disability and severe behavioural problems are often placed outside their local area. Under current regulations, the responsibility for social care remains with the ‘exporting’ local authority. However, the health care responsibility passes on to the ‘importing’ health authority, usually after a period of around 3 months. In parts of the country where a large number of specialist (and often private) residential homes have opened, this results in an unexpected and unplanned increase in the workload of local learning disability/mental health services.

To plan for effective service delivery and appropriate resource allocation, there should be a proper dialogue between the ‘exporting’ and ‘importing’ agencies prior to a placement being made. The spirit of the Department of Health guidance on this issue would suggest that in organising a placement in a residential or nursing home, the local authority must liaise with the district health authority responsible for securing health services for that person (North and Mid Hampshire Health Authority and Hampshire County Council Social Services, 2001). If there are ongoing mental health needs, these should be funded theoretically by payment from the original health authority to the ‘out-of-area’ health authority through a service-level agreement. This should compensate for any extra burden on health services in ‘receiver’ areas (Reference Forsyth and WinterbottomForsyth & Winterbottom, 2002).

Anecdotal evidence would suggest that this is not happening at present. The result is that some health districts take on the out-patient care of people placed within their area after the customary 3-month period, whereas others refuse to do so, insisting that the responsibility should stay with the original health district. Clearly, this is not a satisfactory situation.

No systematic research has been done in this area. There has been some work looking at the extent of out-of-area hospital admissions in general psychiatry (Reference Glover and BindmanGlover & Bindman, 2001), but none looking at residential placements. Within learning disability, there have been no studies that attempt to quantify this problem or suggest practice standards. This baseline survey from Leicestershire addresses the issue by focusing on the catchment area covered by one consultant psychiatrist (total population 230 000; out-patient case-load 220-230).

Because there were no published standards for practice in this area, the authors drew up an initial checklist that was modified later based on feedback from 10 psychiatrists working with community learning disability teams in the Trent region. Feedback was sought also from the community learning disability team for North Leicestershire - a multi-disciplinary body consisting of community nurses, psychologists, outreach team workers, physiotherapists, social workers and occupational therapists. As a result of this wide-ranging discussion, the following practice standards were formulated for use prior to a residential placement being made:

1. 1. Contact and discussion between the health teams involved.

2. 2. Formal hand-over arrangements for psychiatric management.

3. 3. A written summary of the clients:

   1. i psychiatric and medical history;

   2. ii risk assessment/management strategies (it was recognised that this information generally would be contained in a discharge summary).

4. 4. Care Programme Approach arrangements for all patients with mental health problems and learning disability.

5. 5. Negotiations and agreements with the receiving health authority regarding costs for professional input.

Method

The active out-patient case-load of one consultant psychiatrist in learning disability in Leicestershire was surveyed. The period covered was one year from May 2000. People from outside Leicestershire who were placed into residential facilities in the area were identified. Their case-notes were scrutinised and information on socio-demographic and clinical variables was collected.

The ICD-10 (World Health Organization, 1992) diagnostic categories were used throughout. In addition, a category of ‘borderline’ learning disability was used for one individual whose level of cognitive functioning was above that of mild learning disability.

To assess the behavioural and risk profile further, an adapted version of the Overt Aggression Scale (OAS; Reference Yudofsky, Silver and JacksonYudofsky et al, 1986) was used. The OAS scoring was done by the consultant psychiatrist (R.A.) and represented a lifetime score. On the OAS, violent behaviour is divided into the categories of verbal aggression, physical aggression against objects, physical aggression against self and physical aggression against other people. We added the categories of sexually aggressive behaviour and self-neglecting behaviour to this list. Incorporation of the OAS into routine clinical practice as part of a multiaxial classificatory system in learning disability has been described before (Reference Cooray and TyrerCooray & Tyrer, 2000). Using the scale in this way helps to generate a risk profile of the patient group.

Compliance with the suggested practice standards was examined and the data were analysed using the Statistical Package for the Social Sciences (SPSS, Version 6.0).

Results

A total of 29 patients were identified as out-of-area placements, with a mean age of 34 years (s.d.=9, age range 20-59). Twenty-one were male (72%) and eight female (28%). Twenty-eight (97%) were of White Caucasian ethnicity and one (3%) was African-Caribbean. All 29 (100%) were placed into private residential homes within the area and 27 (93%) were known to specialist health services even before placement. The mean length of this contact was 13 years (s.d.=11; range 1-49).

The various clinical diagnoses are summarised in Table 1. On the OAS it was found that 10 patients (35%) exhibited physical aggression against self (self-harming behaviour), 25 (86%) directed violence at others, 25 (86%) directed violence towards property, 6 (21%) had sexually inappropriate behaviour and 25 (86%) were verbally violent towards others. Seven (24%) were considered at significant risk of self-neglect, wandering, etc.

Table 1. Clinical variables - diagnostic category

Degree of learning disability	Borderline	1 (3%)
	Mild	13 (45%)
	Moderate	10 (35%)
	Severe/profound	5 (17%)
Cause of learning disability	Down's syndrome	2 (7%)
	Other	5 (17%)
	Unknown	22 (76%)
Pervasive developmental disorder	Present	12 (41%)
Personality disorder	Present	1 (3%)
Alcohol or illicit drugs (harmful use/dependence)	Present	1 (3%)
Mental illnesses	Schizophrenia and psychosis	3 (10%)
	Recurrent depressive disorder	7 (24%)
	Bipolar disorder	1 (3%)
	Dementia	1 (3%)
	Other	5 (17%)
Behavioural disorder only (those with no mental illness, no pervasive developmental disorder, no personality disorder, no substance abuse)	Present	6 (21%)
Epilepsy	Present	5 (17%)

Regarding professional involvement with patients, the psychiatrist and the community psychiatric nursing team were involved with nearly all the patients: 29 (100%) and 28 (97%), respectively. The assertive outreach team was involved with 12 (41%), clinical psychologists with 9 (31%), speech and language therapists with 5 (17%) and occupational therapists with 2 (7%). Overall, a mean of three professionals from different health disciplines were involved with each patient (s.d.=0.8; range 1-4). In terms of psychiatric outpatient contact, they had a mean of five appointments per year (range 3-8). It was clear that these patients needed significant professional input.

Compliance with each of the proposed practice standards is summarised in Table 2. As expected, compliance was uniformly poor, with well over 90% not achieving any standard.

Table 2. Compliance with proposed practice standards (all standards to be met prior to the residential placement being made)

Practice standard	Standard met
Standard 1	3 (10%)
(Contact and discussion between the health teams involved)
Standard 2	2 (7%)
(Formal hand-over arrangements for psychiatric management)
Standard 3	3 (10%)
(A written summary of the client's psychiatric and medical history, risk assessment and management strategies - usually part of a discharge summary)
Standard 4	0
Care Programme Approach arrangements for all patients with mental health problems and learning disability)
Standard 5	1 (3%)
(Negotiations and agreement with the receiving health authority regarding costs for professional input)

Discussion

Out-of-area placements are often used when higher levels of specialist services are needed, implying greater health needs (Reference Forsyth and WinterbottomForsyth & Winterbottom, 2002). The people who are placed in this manner have complex and varied needs and often a greater incidence of ill health than the rest of the population. Unless service-level agreements are in place, the receiving health authority ends up funding any future mental health needs. In effect, this can mean that the health burden of such placements is particularly high (Reference Forsyth and WinterbottomForsyth & Winterbottom, 2002). It therefore follows that it is essential to have formal meetings between referring and receiving teams before and after transfer of a patient.

This baseline survey and preliminary audit from Leicestershire focuses on people with learning disability placed within one consultant's catchment area from outside the county. It is a small survey involving a selected sample and hence arguably the findings may not be generalisable. However, it addresses an issue that would be familiar to most practising clinicians and this paper is the first attempt to quantify this problem, which for a long time has been the subject of anecdotal (and often anguished) discussion. It stimulates the question of responsibilities of the ‘exporting’ health care authority prior to the transfer of a patient. As is evident from the findings, there is very little contact and discussion from the ‘exporting’ health team. Formal hand-over arrangements are seriously deficient and consultation with specialist health services prior to the placement being made is almost non-existent. Often the first indication about the patient comes from a referral letter from the local general practitioner around three months after the placement has been made; the process of tracing the old history and treatment details starts thereafter. Clearly, this is not the best starting point for formulating a treatment plan. It all seems particularly inappropriate when, as this paper highlights, the vast majority of these patients have been known to the specialist services of their originating districts for many years.

Our data confirm the general belief that people placed in this manner often have high health care needs. In this study, over 40% have an autistic spectrum disorder, 60% have mental illnesses and 80% exhibit significant violence towards others. Needless to say, providing professional input for this group involves a number of disciplines and can be resource intensive, both in terms of labour and time. In this context, early negotiations between the two relevant health authorities regarding costs for professional input assume importance. All 29 patients who were involved in this audit were placed in private residential homes, many of which are offering extremely expensive placements. Perhaps thought should be given to whether some of the cost for the additional professional input should be written into their contracts.

In the area of residential care, it may not be realistic to expect that all areas of the country would have a uniform distribution of specialist skills and resources. Perhaps it is inevitable, therefore, that the practice of placing clients ‘out of area’ will continue, in some shape or form. What can and should improve, however, is the mechanism for ensuring a smooth and seamless transition of health care responsibilities when these placements occur. This mechanism also should address the question of adequate funding for the increased health needs, ‘following the patient’ into the new placement. This is an issue that health authorities need to address urgently. The standards recommended in this paper may provide a starting point for such discussions. As far as individual clinicians are concerned, we feel that these standards would represent a set of minimum goals to ensure that their patients receive the best standard of care. The structure of the Care Programme Approach probably represents the best vehicle for delivering these standards.

Although this study focuses on cross-boundary flow of people with learning disabilities, the issue addressed is equally relevant to anyone with severe mental health problems. Hence, the subject of the study has resonance for all mental health professionals and service users. We have demonstrated that a considerable problem exists in this area and hope that it will encourage others to explore this further.