Case-loads are an important parameter for workers in mental health services because there is a limit to the number of people who can be looked after effectively at any one time. This is now formally recognised; for example, in the recently published National Service Framework for Mental Health (Department of Health, 1999) it states that workforce planning will be carried out to estimate the numbers for each professional discipline in mental health needed in 2002 and 2005. For those practising assertive community treatment (ACT), the specific requirement of no fewer than eight and no more than 12 cases per worker has been formulated (Reference Stein and De SantosStein & De Santos, 1998). For some years in the UK it has also been recommended that community mental health teams (CMHTs) should concentrate on seeing those with severe mental illness and to use other services, particularly those in primary care, for those with other types of psychiatric disorder (Department of Health, 1996, 1998, 1999). However, in addition to this, one of the tenets of the Care Programme Approach (CPA) is that the responsible medical officer (invariably a consultant) takes medical responsibility for all patients attached to the clinical teams of which he is a member. It was decided that the case-load size of each mental health worker in CMHTs in an outer- and an inner-London psychiatric service would be examined and the variation between case-load sizes in different disciplines would be studied. The study was carried out in response to a request from the purchases of the service that the proportion of those with severe mental illness seen by the community team be increased to 70%.

Method

The study was carried out in the areas of Brent and North Westminster, which represent a population of 351 587 (243 466 aged between 15 and 65) served by the North West London Mental Health Trust, a specialist trust for mental illness that was created in 1993 (this has subsequently merged with other trusts forming the Brent, Kensington, Chelsea and Westminister Mental Health NHS Trust).

The definition of severe mental illness for the purpose of this study included all patients with either: (a) any psychotic disorder, (b) a diagnosis of severe depressive illness, (c) a primary diagnosis of severe personality disorder, or (d) treated on the higher level of the CPA.

This shows many similarities with a generally ill-defined group (Reference Schinnar, Rothbard and KanterSchinnar et al, 1990). The information database, Mental Health Connection (Protechnic Computers Ltd, Cambridge CB5 8LA) has been used for patients treated by North West London Mental Health NHS Trust since 1994. This records basic demographic information and ICD-10 diagnosis (World Health Organization, 1992) on all patients seen and is assessed by clinicians within the trust. However, in some instances, particularly if patients have not been seen by a doctor, the ICD-10 diagnosis was not recorded and further information was required. In the first instance all staff were provided with printouts of patients on their case-load and asked to check these, confirming which were being seen currently. Where diagnoses were not recorded, the relevant clinicians were asked to examine case notes to complete these. Second, all junior doctors, who had larger case-loads and the added complication of regular staff rotation, were also asked to check their printed records of case-loads and to remove those whom they did not recognise as actively being seen. However, some patients were seen by other junior or senior doctors, therefore, checks were made to ensure that patients were not double-counted. For both of these groups the criteria were agreed that patients with no contact for 12 months and no out-patient appointment pending, and who had not attended on three occasions and had no appointment pending, would be removed from the case-load. Patients unaccounted for — with a contact within the previous 12 months but no appointment pending — had their case-notes removed from file and scrutinised by the junior doctor if possible and by D. G. if not. This part of the exercise was clinically useful because it also identified those patients who should have had an appointment but through oversight had not been sent one. This exercise showed that 1574 (43%) of the original junior doctor case-load (3666 patients) were not active and further enquiries showed that a further 850 consultant episodes could be closed (Table 1). In Brent 6218 patients were on the case-load (2.55% of the total population at risk), compared with 3986 (5.2% of the population at risk) in Westminster. Overall, 41.5% of the patients seen in both parts of the trust had severe mental illness, considerably less than the 70% postulated by the purchasers. Of the individual professional disciplines, occupational therapists had the smallest case-load but the highest proportion of patients with severe mental illness — much of their work involved the care of people with schizophrenia and other psychosis, often of a chronic nature with functional deficits. The proportions of those with severe mental illness in other disciplines were roughly similar, with consultants having 56% of those with severe mental illness on their case-load.

Table 1.

Mean case-loads of professional disciplines within Brent and North Westminster services of London and average case-loads of whole service between December 1995 and November 1997

Discipline/geographical area	Average case-load per year	Per cent with severe mental illness
Consultant	185-317	66%
Specialist registrar	80	56%
Senior house officer and other junior doctors	98	49%
Community psychiatric nurse	21	55%
Occupational therapist	16	67%
Clinical psychologist	28	49%
North West London (Brent)	6218	41.5%
North Westminster	3986	41.5%

The medical members of the team had much larger case-loads, including many patients attending out-patients clinics who only saw a medical member of staff. Consultants had the highest case-load, with an average of 317 across all consultants within the trust. However, it was appreciated that this figure might be falsely inflated by the requirement for all patients to be registered with a consultant. In order to check the true likely case-load using the criteria described above, a separate examination was made of the case-load of one consultant (P.T.) in one of the sector teams. P.T. was somewhat unusual; carrying both a sector case-load and also being the consultant in charge of the (supra-sector) rehabilitation team, however, the rehabilitation aspects of his work were not considered in this exercise. The assessment was carried out on all patients under his care on 1 December 1996. At that time 419 patients were allocated to him and of these 29 (7%) were mis-allocated because they were under the care of the rehabilitation team, 43 (10%) had seen a junior doctor only, not P.T., and 98 (23%) were on the case-load but had not been seen by P.T. in the previous 12 months. This led to a total active case-load of 249. Assuming that this was representative of other consultants (an assumption that could be challenged), the revised figure for consultant case-load was reduced to between 185 and 317, the range shown in Table 1.

Discussion

The average case-load of non-medical members of the community mental health teams in the North West London Mental Health Trust appears to be appropriate although there is considerable variation between nurses in different settings. This study was confined to CHMTs covering general adult psychiatry and did not include child and adolescent psychiatry, old age psychiatry, substance misuse and special services (e.g. nurses in clozapine clinics). The continuous monitoring of case-load by these disciplines has probably helped to keep the numbers within manageable proportions and it seems likely that an average target of 20 patients per case-load, with approximately half having severe mental illness, is a reasonable expectation for community services. It represents a marked improvement for community nursing staff, who in a previous survey prior to the introduction of CMHTs in 1994 had a median case-load of 62 (Reference Weaver, Patmore and CunninghamWeaver et al, 1999), however, this included a large number of patients who attended for depot injections only. This group constitutes one that is not in the normal sense of the word ‘case managed’ and case-loads of over 100 are not impossible for nurses dealing with these patients (Reference Tyrer, Hawksworth and HobbsTyrer et al, 1990).

The much higher numbers of patients seen by medical staff, who are rarely formally acknowledged as case managers, is a greater concern. It is clear from the number of people not on the active case-load that discharge of patients was not always a high priority for these staff and there was no consistent policy to keep the numbers within manageable limits. The same applied even more strongly to the average consultant case-load. The case-load for consultants appears unmanageable as these numbers cannot be monitored satisfactorily without additional help. At the same time, the consultant, as the designated responsible medical officer, is responsible in the statutory sense for all these patients. Some of the difficulties experienced by consultants in general psychiatry in recent years, often leading to early retirement, suggest that this level of responsibility for such large numbers is not viable in the long term. If consultants are to play an active part in CMHTs and provide a valuable source of clinical expertise, there needs to be a better way of allocating the responsibilities of the responsible medical officer so that greater sharing is achieved and the dangers of the consultants merely becoming bureaucratic administrators of their case-load are avoided.