Abstract
A new community mental health service, the Mobile Assertive Care (MAC) team, offering intensive community based treatment for people with severe and persistent mental illnesses was evaluated twelve months after its commencement. Of the sixty-three clients referred within the first twelve months of operation, twelve-monthly outcome data was available for 54 clients at the time of this study. Results indicated clinically significant reductions in days spent in hospital, and statistically significant improvements in clinical and functional status.
The development of regionalised mental health services in South Australia reflects an increasing emphasis on community based treatment approaches for people with severe mental illnesses [1,2,3]. One component of these services, Mobile Assertive Care teams, are based on the PACT (Program for Assertive Community Treatment) [4] model which originated in the United States to assist clients with severe mental illnesses reintegrate into the community. This model, more recently referred to as assertive community treatment [5,6], is characterised by multidisciplinary teams working with smaller caseloads and targeting care to the most severely mentally ill. Similar services interstate and overseas have demonstrated the efficacy of community based assertive care approaches, particularly in reducing hospitalisation and improving adaptive functioning [7–12].
There are currently four MAC teams operating in metropolitan Adelaide. This evaluation pertains to the Northern MAC team which serves a population of approximately 200 000 in the northern suburbs. Staffing includes two community mental health nurses, two social workers, an occupational therapist, a clinical psychologist and a part time consultant psychiatrist. The team has adopted a clinical case management approach [13,14] to provide treatment and support services to clients with severe mental illnesses who require more intensive support to maintain community tenure. Referral criteria includes a primary diagnosis of severe and persistent mental illness (usually with psychotic features), aged between 18 and 65 years, poor living skills, a recent history of difficulty in engaging in community mental health services and or recurrent psychiatric hospitalisations. Typically, clients are taken on for a minimum of six months and are referred from other components of mental health services The maximum staff to client ratio is 1:10, with the majority of client contact occurring in the client's home. Each team member has detailed knowledge of all clients and has input into their care management plans. This team based approach utilises a range of multidisciplinary skills and targets access to clinical and rehabilitation services and improving independent living skills. The team offers formal psychoeducation programmes for both clients and carers, and facilitates access to a community rehabilitation centre. The team also provides crisis management to clients with support from an extended hours crisis service for after hours assistance.
Assessing the effectiveness of community interventions for sufferers of severe and persistent mental illnesses poses some methodological difficulties in determining what constitutes successful outcomes [15–19]. The protracted nature of severe psychiatric disorders and their heterogeneity in presentation and course mean that for many, treatment may have no endpoint. Thus outcome may signify assessment at one point of an ongoing process of care. Also, the pervasive nature of severe psychiatric disorders means that many aspects of a sufferer's life may be affected, so that a multidimensional approach is required to address the range of social contexts affected by mental illness. The outcome measures adopted for this evaluation reflect the principal MAC service objectives of reducing symptoms, improving adaptive functioning and maintaining community tenure (or reducing the need for hospitalisation). The applied nature of this evaluation precluded the use of a control group, thus limiting the identification of significant service factors contributing to change. However, the absence of randomised control groups in applied settings should not restrict clinicians in their endeavours to provide evidence of service outcomes [20].
METHOD
Using a repeated measures longitudinal research design, all MAC clients were assessed following referral to MAC and at six monthly intervals thereafter. Assessment instruments included measures of clinical symptoms (Brief Psychiatric Rating Scale) [21,22] and adaptive functioning (Life Skills Profile) [23]. Hospital admission rates (number of days hospitalised) were obtained from client records for the two years prior to MAC referral and the twelve months post-referral. Admission data were collated for each six monthly period over the three years to compare hospital admission patterns pre- and post-referral.
The Brief Psychiatric Rating Scale (BPRS) consists of an 18-item semi-structured interview to assess psychiatric symptoms with symptom severity measured on a 1 to 7 Likert scale. The Life Skills Profile (LSP) is a 39-item measure developed specifically to assess functional disability associated with severe and persistent mental illness. Ability is scored on a Likert scale from 1 to 4 on five functional dimensions of self-care, non-turbulence, social contact, communication and responsibility. Both scales have adequate psychometric properties.
RESULTS
At the time of this study twelve-monthly outcome data were available for 54 clients. Of these, 38 had received twelve months of MAC services and 16 had been discharged to less intensive community services. Outcome data were unavailable for two clients who had left the area. There were 38 males and 16 females with a mean age of 30.9 years (SD=9.6), 52 were single and two married. Forty-two clients lived alone (predominantly in government housing) and 12 resided with their families (usually with parents). Forty-eight clients had a primary diagnosis of schizophrenia and six had a diagnosis of major affective disorder. Fourteen clients were under legally enforceable community psychiatric treatment orders.
There were clinically significant reductions in the number of days spent in hospital, per six months, for the twelve months post referral when compared to hospitalisation rates for the two years prior to MAC referral (see Table 1). Statistically significant improvements were found in the five LSP scales of self care, non-turbulence, social contact, communication and responsibility (see Table 2), and in clinical functioning as measured by the BPRS (see Table 3). There were also statistically significant improvements in discharged clients (N=16) total LSP (t = −2.5, p<0.05) and BPRS (t = 3.9, p < 0.001) outcome scores when compared with current MAC clients (N=38).
Days spent in hospital per six monthly period pre and post MAC intervention
Differences between Life Skills Profile scores at 1 (pre MAC) and 2 (12 months post MAC)
Differences between Brief psychiatric Rating Scale scores at 1 (pre MAC) and 2 (12 months post MAC)
DISCUSSION
The purpose of this evaluation was to identify whether the MAC service objectives of reducing illness symptoms and improving adaptive functioning and community tenure had been met. The applied nature of the study limited the research design due to the necessity for MAC to provide services to all accepted referrals. A repeated measures design which included a randomised control group of clients meeting MAC referral criteria who received less intensive community follow-up would have enabled a more conclusive analysis of which components of MAC influenced outcome. Ongoing data collection, including that from discharged MAC clients, is required to determine if positive outcomes are maintained. Methodological problems inherent in longitudinal designs such as the effect of time itself on outcome, regression towards the mean, spontaneous remission and the variations in baseline levels of functioning affecting individual rates of change are to be considered in this evaluation. Also, the use of additional outcome measures such as quality of life are required to clarify the clinical significance of the statistically significant improvements found in clinical and functional status.
The finding that hospitalisation was significantly reduced following the introduction of MAC is consistent with other studies of similar services in Australia and overseas [3,6–10]. However, data collection over a longer time period is necessary for a more conclusive analysis as it would be expected that high hospitalisation rates would precede MAC referral and longer time frames are required to capture illness relapse patterns. Longitudinal readmission data from clients discharged from MAC is also required to see if relapse occurs more frequently among those referred to less intensive services. Important considerations in using rehospitalisation rates as an outcome measure include the risk of devaluing the significant role hospitals play in client care and the potential biasing factors in interpreting data from services where there may be an over-emphasis on avoiding hospitalisation. MAC considered alternative interventions to hospitalisation if clients were demonstrating early warning signs of becoming unwell, such as increasing support or medication adjustment, but did not compromise attention to risk to clients if hospitalisation was needed. Further criticism of the use of rehospitalisation rates as an evaluation measure includes the argument that decisions to hospitalise may be more reflective of local ideological, political and economic factors rather than client needs [13]. However, rehospitalisation rates remain an important outcome variable for consideration if a major goal of community services is to provide secure community tenure as an alternative to long periods of hospitalisation
Further research is required to establish what factors specific to MAC services contribute to improved outcomes. The two characteristics unique to assertive care models of care are a relatively low client to staff ratio and services targeted to those with severe and persistent mental illnesses. The additional time available for clients and the specialisation of services appear significant contributing factors. Although smaller caseloads are not an option for many community services, the use of small multidisciplinary community teams with discrete roles and targeting specific client groups enables staff to clarify their service objectives and expected outcomes. These factors enhance opportunities for clinicians and treatment services to incorporate evaluation and quality assurance activities into their everyday clinical practice. The absence of scientifically controlled conditions, such as randomised control groups, should not preclude the measurement by clinicians of treatment outcomes. Treatment services operate in applied settings and therefore need to balance scientific rigour with the necessity to provide evidence-based treatment practice.