A survey of liaison psychiatry services in general hospitals and accident and emergency departments: do we have the balance right?

AIMS AND METHOD By use of a telephone survey, we aimed to investigate liaison psychiatry 
services of all 29 general hospitals in Greater London. We specifically 
enquired about services to accident and emergency (A&E) departments. RESULTS We identified wide variations in staffing, working hours and patient groups 
seen. Fourteen services (48%) worked over 24 h and 4 (14%) had specific 
A&E teams. Twelve services (41%) had established or planned working links 
with community crisis services. CLINICAL IMPLICATIONS Generally staff numbers fell below national recommendations and there were 
frequent gaps in service provision. The recent focus on emergency care has 
lead to an increase in A&E services, but there is a risk that liaison 
psychiatry services for other general hospital patients are being 
neglected.


TA M S I N K E W L E Y A N D J I M B O LT O N
A survey of liaison psychiatry services in general hospitals and accident and emergency departments: do we have the balance right?

AIMS AND METHOD
By use of a telephone survey, we aimed to investigate liaison psychiatry services of all 29 general hospitals in Greater London. We specifically enquired about services to accident and emergency (A&E) departments.

RESULTS
We identified wide variations in staffing, working hours and patient groups seen. Fourteen services (48%) worked over 24 h and 4 (14%) had specific A&E teams. Twelve services (41%) had established or planned working links with community crisis services.

CLINICAL IMPLICATIONS
Generally staff numbers fell below national recommendations and there were frequent gaps in service provision. The recent focus on emergency care has lead to an increase in A&E services, but there is a risk that liaison psychiatry services for other general hospital patients are being neglected.
Although the number of liaison psychiatry services in the UK is increasing, their development is idiosyncratic and services often fail to meet the recommendations of the Royal Colleges of Physicians & Psychiatrists (2003). Across the UK there is a wide variation in staffing and service delivery (Howe et al, 2003;Ruddy & House, 2003;Swift & Guthrie, 2003).
Recent English health service initiatives have led to an increased focus on emergency mental healthcare, which potentially influences liaison psychiatry provision to accident and emergency (A&E) departments. The Department of Health (2001) has recently set standards to reduce patients' attendance times in A&E departments. In addition, the National Service Framework for Mental Health (Department of Health, 1999) has required specific services to be established for patients in crisis, many of who will attend A&E departments. However, there is no optimum model of psychiatric service delivery to A&E. Also, there is a risk that A&E mental health services develop at the expense of other hospital departments.
As a city, London is unique in the UK in terms of its size, ethnic diversity and organisation of health services. As part of the establishment of a network of liaison psychiatry services in Greater London, we surveyed the current service provision to all the district general hospitals. We aimed to investigate the staffing and service provision of each service and to enquire about service developments, particularly with regard to A&E departments.

Method
Greater London comprises the 32 London boroughs and the City of London. Using information supplied by the Department of Health, 29 Greater London district general hospitals with A&E departments were identified. Information on bed numbers was obtained from hospital websites or directly from bed managers.
During the second quarter of 2004, a telephone survey was carried out using a predetermined list of questions. In each case, we tried to speak to a wellestablished member of the liaison psychiatry team. We enquired about the number of clinical team members and their professions. Specialist registrars (SpRs) in psychiatry were not included in these figures, as such posts are supernumerary and may not continue beyond the current trainee's attachment. We established details of service delivery. Hours of work were categorised into services operating within core 'working hours' (09.00 to 17.00 h, Monday to Friday), those delivering an extended-hours service and those running 24 h per day. The survey enquired about specific patient groups seen and whether teams were funded and managed by a mental health or an acute trust. We also enquired whether the hospital had a mental health in-patient unit on site. Efforts were made to establish whether there were trends in service development by asking about recent service changes and teams' priorities for future development.
Data were analysed using the Statistical Package for the Social Sciences version 11.5 for Windows. Staffing levels were compared with the joint Royal College recommendations (Royal Colleges of Physicians & Psychiatrists, 2003).

Results
Information was obtained from all 29 district general hospitals. Four hospitals had separate liaison psychiatry teams covering either A&E or the rest of the hospital. In such instances, data from these two teams were combined in order to compare services between hospitals. Table 1 gives a profile of the different services and  Table 2 summarises their staffing levels.

Bed numbers
The mean number of in-patient beds for the hospitals surveyed was 635 (range 257-1267, s.d.=228).

Working hours
Of the ten services working extended hours, seven included weekends and three operated on weekdays only.

Staffing
Overall, the size of teams varied between 1 and 25 whole-time equivalent staff (mean=8.2, s.d.=5.9). Nine teams (31%) consisted of nursing staff only. Ten services (34%) had no dedicated consultant psychiatry sessions, although these teams usually mentioned that they could contact a psychiatrist for advice.

Patient groups
All teams provided a service to their A&E department and assessed patients following episodes of self-harm. Five services (17%) operated in A&E departments only. Patient groups seen are listed in Table 1.

Services for older adults
Nineteen services (66%) accepted referrals of older adults, although 3 of these teams worked only in the A&E department. Several teams mentioned that community psychiatry services for older adults visited the hospital to see patients. Four on-site liaison psychiatry services specifically for older adults were identified, for which limited data were collected. Each had a half-time consultant with one to two additional nursing or junior medical staff.

Funding and management
Twenty-five services (86%) were managed by the mental health trust, 3 (10%) by the acute trust and 1 (3%) was original papers Kewley & Bolton Liaison psychiatry services  jointly managed. Fifteen services (52%) were funded by the mental health trust, 10 (35%) were jointly funded and 4 (14%) were funded by the acute trust.

On-site psychiatric unit
Eighteen district general hospitals (62%) also had a psychiatric unit on site. There was a significant association between the presence of an on-site unit and the liaison psychiatry team delivering a 24-h service (w 2 =6.43, P=0.01).

Service developments
Those teams providing extended and 24-h services generally reported that they had increased their hours of service during the previous 2 years. Three teams (10%) reported that they also delivered the local mental health crisis or home treatment service. A further 9 services (31%) reported plans to establish closer links between liaison psychiatry and community crisis services. The most frequently cited priority for service development was for more staff, mentioned by 24 services (83%). However, only 8 teams (28%) had plans to expand. The second most common priority was for more accommodation, which was mentioned by 10 services (35%).

College recommendations
The joint recommendations of the Royal Colleges of Physicians & Psychiatrists (2003) for a liaison psychiatry service are based on a 09.00 to 17.00 h service in an average-sized district general hospital. Such a service should include one full-time consultant, a senior house officer, 5 nursing staff and 1-2 psychologists.
It is difficult to compare the recommendations with services delivering extended-hours and crisis services. However, only 13 teams (45%) had a full-time equivalent consultant and 17 (59%) had a full-time junior doctor. Only 14 services (48%) had 5 or more full-time equivalent nursing staff; 13 of these delivered a 24-h service and 1 an extended-hours service. Only 1 team had more than 1 full-time equivalent psychologist. Overall, only 1 extended-hours service had all of the recommended components.

Discussion
This is the first survey to investigate liaison psychiatry provision to London's district general hospitals. It identifies wide variations in staffing, hours of work and patient groups seen. All except one service fell short of the recommendations for service provision. However, there has been a recent expansion in services to A&E departments.
Many teams reported that changes in service provision to A&E departments had occurred as a result of recent government targets. A minority of teams also delivered community crisis and home treatment services, work that falls outside the generally accepted remit of a liaison psychiatry service and complicates the interpretation of the survey data.
The recent government targets for emergency mental health provision can be considered an opportunity to bring mental health resources into the district general hospitals. However, there is also a danger that the development of A&E services will be at the expense of services for other general hospital patients.
There is no recommended model for psychiatry service provision to A&E departments. Further qualitative research might help to establish recommendations for A&E services and how emergency mental health services in general should be delivered.
The London population has a number of characteristics that, although not unique among urban areas in the UK, contribute to relatively high levels of mental illness and an associated need for services (Mayor of London, 2003). These include large refugee and minority ethnic populations, high levels of homelessness and single occupancy households, and the fact that London contains many of the most deprived areas in the UK.
Compared with data from previous regional surveys in the UK (Howe et al, 2003;Ruddy & House, 2003), a higher proportion of London district general hospitals had some form of liaison psychiatry service and a higher proportion of services had dedicated consultant psychiatry sessions. However, the combined picture is one of unmet need and a lack of rational planning of services.

Limitations
This study is likely to underestimate the provision of mental health services for general hospital patients, as a number of possible psychiatric and psychological services were not surveyed. Community psychiatric services for children, older adults and substance misuse may provide a consultation service to their local hospital. In addition, individual hospital departments may also employ therapists and psychologists to work with individual patient groups.
An underestimate of services may also have been a result of the omission of SpRs from the survey. Although SpRs are supernumerary and posts are not necessarily filled every year, where present SpRs make a significant contribution to service provision.
We compared each service with the recommendations for an average district general hospital. However, the needs for psychological services of individual hospitals will vary depending on bed numbers, the workload of the A&E department and the epidemiology of the local population. The Colleges' recommendations also point out that larger teaching hospitals with tertiary referral centres will require a larger liaison psychiatry team, including at least two full-time consultant psychiatrists.

Conclusions
This survey describes the wide variety in liaison psychiatry provision to London's district general hospitals, with services generally falling below the recommended standards. Although the focus on A&E provides an opportunity to increase liaison psychiatry services for one hospital department, there is a risk that the needs of other patient groups will be neglected. When services are planned the whole hospital should be considered to ensure that all patients receive comprehensive care.

Declaration of interest
None.