Three months in the life of a community mental health team

cent of new referrals met the priority group criteria, allaying anxieties that community services unceasingly get drawn to the care of those with lesssevere disorders. Furthermore, at 12 months the service had maintained contact with all patients previously admitted to hospital and all those presenting with psychotic disorders (27% referrals). Some changes in service structure are also suggested as a result of this evaluation.

This paper represents a three month audit of a mental health service recently described by Scott et al (1992) which is provided entirely within the community (including the admission facility).The project was undertaken to see if this service could avoid the problem noted by other commu nity services of being drawn away from the care of those with more severe disorders.The service is provided to 50,000 people in North Tyneside (ranking 58th on the Jarman indices) extending from deprived inner city wards to more rural ex-mining communities.At the time of the audit, the community team comprised a consultant/ senior lecturer, a lecturer, 3.5 community psychiatric nurses (CPNs).two occupational therapists (OTs), a senior house officer (SHO), a social worker (SW) and administrative support.The admission unit had an establishment of 12.5 nurses for the nine in-patient beds and the partial hospitalisation programme (Scott et al, 1992).The service currently receives 430 to 520 referrals per year.
The service primarily aims to provide an asser tive outreach programme (AOP) for those with long-term and severe difficulties, with a second ary goal of providing a home-based early inter vention service (EIS) for new referrals (with early return to the GP when possible).The E1S aims to see emergency referrals within 24 hours, urgent referrals within 72 hours and routine referrals within seven days.The co-existence of the AOP and EIS appears to allow the service to run safely with relatively low bed numbers provided in a 'domestic' environment.
A recent 'census' demonstrated that 66% of clients in contact with the service had psychotic or affective disorders, that those with the lowest global levels of func tioning were receiving the most intensive input and that 80% of the CPN and 90% of the medical case-load comprised those with long-term and severe disorders.However, a census cannot pro vide a total picture of the clinical work of the team.It does not give information regarding referral patterns nor change in case-load over time.To do this a three month audit of referrals combined with a one year follow-up of these cases was undertaken.The specific aims of this audit were: to assess the efficacy of the prioritisation process through a review of new referrals and patterns of contact with the service of indi viduals with severe and enduring mental illness, and to assess the efficiency of the home-based EIS.

The study
Over a three month period the case-notes of all new referrals to the community team were reviewed by an independent assessor (LS).A standardised pro forma was used to record refer ral information, demographic characteristics and DSM-III-R diagnosis.Treatment setting, length of contact with the service and the professionals involved in assessment or treatment of the client were noted.Where individuals were not seen, a probable DSM-III-R diagnosis was made from referral information.

Referral patterns
One hundred new referrals were received during the three month study period.This is nonsignificantly less than the usual quarterly refer ral rate to the service (median 125) (Smith, 1993).The majority of referrals (66%) were from GPs, only 5% were self-referrals.Fourteen cases were re-referrals.Only 30% of individuals were married or cohabiting.

Assessment patterns
All those with psychotic illness, 88% of those with depressive disorders and 90% of those with a past psychiatric history were assessed.How ever, 17 referrals were not seen.While five were directed to more appropriate services (e.g.psychogeriatrics; Relate), 12 individuals declined assessment for unspecified reasons.Probable diagnoses were: depressive disorder (n=2), anxiety/adjustment disorders (n=4), eating dis order (n=l), personality disorder (n=3), unknown (n=2).The team made one to three attempts to establish contact before returning care to the GP.
In those seen (n=83), 80% of first assessment interviews were undertaken in the client's home.Ten of 13 emergency and urgent referrals were seen within the specified target times and six of these cases were admitted to hospital within 36 hours.In the three other urgently referred cases the clients chose to defer the appointment offered (one was going on holiday!).Only 40% of routine referrals were seen within the target time of one week, although 70% were seen within 14 days.The median time between referral and assessment was 12.5 days.

Psychiatric history
Ten per cent of referrals were newly diagnosed cases of psychosis (schizophrenia=7), 34% had depressive disorders, 30% had other persistent disorders (e.g.eating disorder, personality dis order, drug/alcohol problems), only 26% had anxiety, adjustment or other less severe dis orders.Thirteen individuals had an Axis II and five had a co-existing Axis III disorder.Sixty people had a past history of mental health problems, the majority having been treated by their GP.

Treatment setting
After assessment, 12 clients were referred back to the GP with advice on management and four were referred to specialist psychiatric services.Of the remainder, 47 patients received home treatment, two entered the partial hospitalis ation programme and eight became in-patients (five with affective disorders, one with schizo phrenia, one with schizoaffective disorder, one with anorexia).All were initially voluntary admis sions although two were subsequently detained under the Mental Health Act.The median length of in-patient stay was 24 days.

Service contact
At 12 months, 27 patients (23 of whom were single) were continuing to receive treatment and follow-up.This group included all those individu als with psychosis and all those who had been admitted.Only three of the 27 patients had anxiety/adjustment disorders.Of the 73 discharged cases, the median length of time registered with the service was 16 weeks (1-38).
Overall, 21 people were treated by more than one professional.Medical input was offered to 68 patients at some time during their contact with the service.Two cases who had originally been referred but not seen were referred at a later time.

Discussion
The development of locality based services for mentally ill people has been the subject of increased scrutiny by politicians, policy makers and the public.One of the most important requirements of new service models is that resources are targeted at the most vulnerable individuals with severe, long-term disabilities.Continuous monitoring of service developments are required to ensure that this priority is main tained.
This audit represents an evaluation of the utilisation of resources for new referrals to a community-based service.Although the number of referrals was slightly less than usual, some general conclusions can be drawn.About 75% of referrals to this service are for severe or enduring disorders: 10% were for psychotic disorders, a further third were for depressive illnesses and a further third were for patients whom GPs often find difficult to treat in the community (e.g.those with personality and eating disorders); over half had a past psychiatric history; the majority were socially isolated and 70% required input from a psychiatrist on at least one occasion.Only one in ten of those seen (8% overall) was admitted, but all of the psychotic patients and about one in three of those with depressive or other disorders were still in contact with the service one year after referral.These findings offer some support for the efficacy of the service in primarily offering community-based treatment, in communicating about its priority client groups to the referring agencies and in maintaining contact through the AOP with this target population.
One of the obvious negative findings of this survey is that 12% of referrals were not seen despite being offered home assessment.None of the refusers had a definite psychotic illness, but other reliable information is lacking.The total failure to access ten clients (two were assessed on further referral) may reflect their own reluc tance to use the service.Discussions are under way with referring agencies to determine whether there are any common characteristics in such cases.Given that home assessment is an option for all clients and that one to three alternative appointments were offered, it is not clear what other changes can readily be introduced.Do the mental heath services need to be more active in trying to access this group or do the referrers need to offer the individuals more information regarding reasons for referral?
While the emergency and urgent referral com ponents of the EIS appeared to work efficiently, it was not possible to achieve the target of complet ing all new assessments within a week.Given the catchment area characteristics, the relatively small size of the community team and their ongoing caseloads, this was not unexpected.While undertaking 40% of routine assessments within the target time (and 70% within 14 days) was better than predicted, the question arises as to whether persisting with this goal is appropri ate.The philosophy underpinning the EIS is that early intensive contact may mean people are assessed and treated before the disorder has led to a further deterioration in their ability to func tion.Theoretically, this may avert the need for admission to hospital or allow earlier treatment and stabilisation of the condition with the possi bility of a more rapid discharge back to the care of the referrer.The lower admission rate, lower section rate and shorter median time to dis charge (16 v. 28 weeks) compared with other catchment areas in the district (Smith, 1993) initially appear to support this view.However, those with the most severe disorders or in crisis (e.g. a suicide risk) were seen mainly through the emergency or urgent referral process and in half of these cases early intervention did not avoid admission.Furthermore, in a third of all cases early discharge was not deemed appropriate.
From the above it is clear that the effect of the EIS on the outcome in the routine referrals is difficult to evaluate; it needs to be investigated through a comparative study with a service employing a different model.Such a prospective study is now underway.Both teams will monitor the time between referral and routine assess ments and information will be gathered from the services on whether patients waiting longer for routine assessment show a deterioration in functioning or if a delay in assessment leads to increased or extended admissions.
Although contact with the community service is relatively brief, it was noted that one in seven of the non-psychotic cases seen in this audit were re-referrals.Previous early discharge may have been detrimental to some patients with apparently less severe illnesses.A brief review of these cases does suggest that some of the re-referrals were predictable (e.g. because of social isolation), and the clients may have ben efited from a more extended contact.However, in others re-referral occurred because the clients were reluctant to access the services except at times of crisis.
In summary, it appears that a community orientated service can achieve its goal of target ing resources at those with severe or difficult to treat disorders.However, the data from this ser vice suggest that at least 10% of referrals are resistant to using the service and changes may be needed to improve accessibility or acceptabil ity to this group.The service was not able to reach its own target of assessing all cases within one week of referral and it is necessary to review the three components of the EIS.As an increase in the staff available is not possible, it is deemed more appropriate to change the target time for routine referrals for two reasons.First, the option of prioritising assessments through the emergency or urgent referral process means that delays do not appear to adversely effect those with severe acute disorders.Second, this audit found that at one year follow-up, about 30% of the cases referred are justifiably (on the basis of the severe and persistent nature of their prob lems) in continued contact with the service.If resources were channelled into the EIS to retain the seven day assessment target there would need to be clear evidence of a positive effect on the outcome of the less severe cases without any detrimental effect on the AOP for those with long-term severe illnesses.The accumulation of clients with long-term difficulties on the case load was anticipated but will need to be moni tored separately to evaluate whether the staff available can retain this priority service at its current level.Three months in the life of a community mental health team