Abstract
Fifteen percent of Australians live in population clusters of fewer than 1000 people [1]. This is defined as a ‘rural locality’ by the Australian Bureau of Statistics. Those responsible for health service planning in Australia have always grappled with the difficult issue of service provision to these small geographically isolated towns. Projected health plans have highlighted the special needs of the geographically isolated [2]. It has been recognised that rural mental health services need to be available (to exist and have sufficient personnel to provide them), to be accessible (the individual client has access to the service) and to be acceptable (to be offered in a manner congruent with local values using a mode suitable for the rural setting) [3]. Health services to isolated towns have traditionally been provided by lone general practitioners with the assistance of small country hospitals. Specialist services have generally been provided in regional centres (with some outreach services usually by non-medical staff).
Some isolated towns have received visiting medical specialists for 20 years or more but there does not appear to have been a structured approach to the allocation of these medical resources. The Rural Aerial Health Service was trialled in New South Wales (NSW) in 1974 with a brief to supply specialised liaison health services to rural general practitioners and secondarily direct patient consultations. Geriatric medicine and mental health services were to be a prime focus. Rural area health services thought to be disadvantaged were targeted. The original plan included the hope that this service would provide some skills to local health staff to enable them to continue the service autonomously. Almost 20 years on, with financial restrictions, the primary focus has become individual patient consultation, rather than local service development [Ambrose G, Rural Aerial Health Service, NSW: personal communication].
The impetus for this project stemmed from 5 years' experience as a visiting psychiatrist to Brewarrina. It became clear that clinical service provision, while valuable in itself, did little to further the standards of psychiatric practice among local health staff. The timing of such visits often did not match emergency presentations of psychiatric disorders and staff sometimes felt at a loss to deal with common acute psychiatric presentations.
The positive impact of improving general practitioners' skills in mental health practice has been established in urban settings, using skills-training programs [4] and liaison attachment models [5]. This project combined such strategies (skills development and joint patient care) to strengthen peer support to health workers, provide psychiatric care to patients in their own setting and to improve the mental health skills of health workers including general practitioners.
Method
Towns in far west NSW were identified in consultation with regional mental health services. Towns were chosen for the presence of a lone mental health worker or being a distant outpost of a visiting mental health worker. The project manager personally visited each possible location, visiting district hospitals, Aboriginal Health Services and general practitioners canvassing interest in the project. Mental health care was identified by these groups as an important need. Health staff requested very practical education sessions focused on common skills in mental health practice.
Mental health skills development sessions
Education evaluation
Evaluation tools were developed that encompassed demographic and training data, knowledge of common areas in mental health practice, application of that knowledge using two clinical vignettes (of an overdose and example of sudden aggression) and attitudes to working with clients with mental health problems. Participant attitudes to working with clients with mental illness were rated using statements adapted from the Patient Desirability Questionnaire [6]. These were completed at the beginning and conclusion of the 10 sessions. Each teaching session was also evaluated by participants. (Evaluation tools are available from the corresponding author.) Indirect evaluation measures included admission rates from each town to the regional psychiatric hospital and psychotropic prescription patterns identified by postcode from the Pharmaceutical Benefits Scheme data (for 18 months prior to and during the project). Evaluation measures were analysed using the Pearson Chi-squared test (for comparison of pre- and post-independent groups) and the Proportions test to examine changes in proportions of psychotropic scripts averaged across project sites in comparison to NSW data. Ninety-five percent confidence intervals were calculated using the normal approximation method.
Clinic visits
The project was promoted in a number of mental health venues seeking expressions of interest from clinicians willing to make an 18-month commitment to a given town. Volunteers were paired (a psychiatrist with another mental health professional) and attached to a given town. Information was sent to the district hospital and general practitioners regarding the ‘visiting team’ including a brief curriculum vitae.
Visits were organised using a 1-day format. Teams flew by light plane charter direct to the service town. Clinics were organised by a local contact person on the basis of referral from any person and usually conducted from 09:00 h to 13:00 h. Teaching sessions were held in the afternoon for 2 h. One team (visiting Cobar) flew by commercial flight, staying 48 h and allowing the clinic to run over to the next day if required.
Direct clinical services were provided to clients in the service town and surrounds. The towns visited were Bourke, Brewarrina, Cobar, Coonabarabran, Coonamble, Lightning Ridge, Nyngan and Warren. If required, referrals were made to the regional centre general hospital psychiatry unit (Dubbo Hospital Special Care Suite), 5 h away, or to the regional psychiatric hospital (Bloomfield Psychiatric Hospital), 8 h away. Tertiary referrals could be made to metropolitan Sydney hospitals. Clinical services were integrated with that provided by regional visiting mental health staff.
Transport costs were met by the project and are detailed in Results. Salaried staff were ‘donated’ to the project by their own area health service. Other project members billed Medicare for services rendered. Salary and Medicare costs have not been included in the cost analysis.
Methods
Results are presented as: (i) direct measures (clinical data, education session evaluation and pre and post knowledge, skills assessment and staff attitudes); and (ii) indirect measures (admission rates and prescription data). Costs are then detailed.
Direct measures
Clinical data
In total, 394 clients (female 62%, male 38%) were reviewed by the visiting teams. Clients were aged between 18 and 30 (n=110, 29%), 31 and 40 (n=84, 21%), over 60 (n=67, 17%), 41 and 50 (n=60, 15%), 51 and 60 (n=51, 13%) or less than 18 (n=22, 5%). Referrals were largely made by general practitioners (70%), although community nurses referred 20% of clients seen. Common diagnoses were major depressive disorder (n=178, 45%), schizophrenia (n=75, 19%), generalised anxiety disorder (n=29, 7%), mental disorders due to a general medical condition (n=23, 6%), bipolar disorders (n=20, 5%), personality disorder (n=19, 5%) or other psychiatric diagnoses (n=35, 9%). No psychiatric diagnosis was made for 15 clients (4%). One-third of clients were seen on more than one occasion. One-quarter of clients were indigenous Australians. Most consultations occurred in the presence of a local health worker: mental health worker, community nurse, Aboriginal health worker or general practitioner.
Education session evaluation
Ten education sessions were completed and targeted common mental health skills. There were 394 attendances by 123 individuals, of whom 85% were female. The individual mean attendance was 3.2 sessions. Only 46% of attendees reported a fair understanding of the topic prior to a session, while 62% reported a good understanding after the session. Most thought information presented was appropriate (97%) and that the skills taught were appropriate (95%). All attendees felt the sessions had some relevance to their work with 50% rating them as very relevant. Most reported they learnt something new at the session (52%).
Knowledge assessment
Prior to the teaching program, 104 women (85%) and 19 men (15%) completed the tool. Most had no mental health training or experience (103, 83%), few had ‘on the job’ experience (7, 6%), while some reported diploma or degree level training (13, 11%). Most attendees were nurses (78, 64%), Aboriginal health workers (10, 8%), general practitioners (9, 7%), teachers (3, 2%) or of other professional background (21, 19%). Other professions included police, ambulance officers, allied health workers (occupational therapists, psychologists, drug and alcohol or domestic violence workers) and administrative staff. Most attendees had some contact with clients with mental illness (77, 63%), while others had regular contact (32, 26%) or no contact (14, 11%). ‘Textbook’ knowledge (as contrasted with practical skills) regarding mental health was assessed using a series of statements requiring a true/false response, such as ‘Major depression effects 20% of people in their lifetime’. Most statements were answered correctly both before and after teaching. Following the 10 sessions, 34 participants completed the knowledge and skills assessments (only eight had completed the tool prior to commencement of the teaching program). Of the eight participants who completed the questionnaire both before and after the 10 sessions no change in response was of statistical significance (using the Chi-squared test).
Skills assessment
Two clinical vignettes were used to assess application of knowledge in a clinical setting. The first vignette was of sudden aggression based on delusional thinking. Most participants (96%) felt a psychiatric diagnosis was warranted but few (10% pre teaching, 23% post teaching) recognised the delusional content. Few participants (11% pre teaching, 38% post teaching) correctly suggested a psychiatric assessment as part of the management. The second vignette was about a young woman in crisis who took an overdose. Most participants could identify a number of stressors contributing to this behaviour and were less likely to identify a major psychiatric disorder as underlying the behaviour (pre teaching 55%, post teaching 45%). More participants were mindful of issues of motivation in suicidal behaviour after the teaching program.
Staff attitudes
Six statements about client desirability were rated on a five-point scale (strongly agree to strongly disagree) like ‘As a professional, I would like to have mentally ill patients as part of my practice’. Respondents were more willing to work with clients with mental illness and less judgemental about their behaviour after the teaching.
Indirect measures
Admission rate
Admission rate to the regional psychiatric hospital before and during the project were examined, finding an increase in admissions from 35 admissions in the 18 months prior and 60 admissions during the same period of the project.
Prescription data
Comparison of psychotropic prescriptions before and during project using the proportions test (proportion of total scripts written) from Pharmceutical Benefits Scheme data
Costs
Project costs included travel ($105 598), part-time research officer ($43 749), equipment ($4029) and sundries ($2874). The travel costs were largely for chartered air transport. One team travelled by commercial airline but costs were similar when accommodation and meals (necessary for the sparse schedule) were included. Visiting psychiatrists and mental health staff were either paid by their respective metropolitan area health service (donating time in return for the staff development gained) or funded by Medicare. These costs were not included. Costs were assigned equally to clinical and educational services, therefore each client contact and education attendance had $198 of associated costs ($134 in travel alone).
Discussion
There is a paucity of data examining models of psychiatric service provision in rural and remote Australia. Rural general practitioners have primarily declared a preference for surgery based outpatient clinics (rather than domiciliary support as preferred by urban general practitioners) [7]. Yellowlees et al. examined psychiatric service delivery, documenting community mental health service expansion, high demands and the development of designated psychiatric beds in Broken Hill within the Far West Mental Health Service [8–10]. Although a common method of service provision, no work was identified evaluating a visiting psychiatric service to rural and remote communities.
This evaluation has shown that some resources can be freed up from metropolitan-based services and temporarily relocated to the country, meeting some of the criteria of an effective mental health service (i.e. availability and accessibility). Further research will examine the third aspect of service: the acceptability of the service to the people using that service. In 10 visits, 394 client contacts were made. These were services largely built from scratch with consultation numbers growing over time. One-quarter of clients were Aboriginal and Torres Strait Islander people, reflecting the general population demographic of the area serviced. Some continuity was provided in that one-third of clients were seen on more than one occasion. Referral data were skewed by the Medicare requirement that clients be referred from a doctor; many referrals started from other sources but were recorded as a general practitioner referral to facilitate bulk billing. Most consultations occurred in the presence of a local health worker promoting skill development as well as continuity of care. Diagnostic details show a preponderance of major mental illness in keeping with other community mental health samples. It may be that clinical services would be more effectively delivered by staying several days in a centre (or visiting a number of nearby towns each visit). While that model had financial appeal, there was little enthusiasm from practitioners for that degree of disruption of their metropolitan lives.
There were some minor teething difficulties in organising effective communication with each of the remote locations. These were overcome with the appointment of local contacts who were keen to liaise regarding the project. The clinical teams visiting each town met on a number of occasions to discuss problems and many commented on a need to shift mind set from a metropolitan-style practice with a wide array of backup resources to that suitable to a rural or remote community with few referral sources. This project brought a challenge for some visiting clinicians to shift from a primarily medical model to a team approach to assessment and treatment of patients. Some staff found the time pressures difficult and managed to cope better in distributing work as the project progressed.
The education packages were well received with largely positive evaluations. The attending groups were of diverse backgrounds, leading to difficulties in meeting all needs. General practitioners participated differently across the sites. It seemed that towns with multiple general practitioners had more involvement in the education sessions. Work pressure for solo general practitioners may simply have hampered involvement. This is a problem as solo general practitioners have difficulty accessing regionalised doctor-focused education due to the unavailability of locum staff. Solo general practitioners also may have found it difficult to demonstrate their skills (or lack of them) in front of non-medical health workers from the same town.
The evaluation tools were of limited use. There was enormous turnover in rural health staff throughout the project, leading to few staff being available to complete the post measure. While sessions were designed to be valuable alone, there was ideally progressive education across the package. The average attendance was, however, only 3.2 sessions.
The indirect evaluation measures were also problematic. Admission and prescription data were both likely to have a lag effect in any real change. Improved case finding by the visiting teams may well have contributed to the rise in admission rates. This may settle over time (with improved local mental health services) or may remain high due to the increased awareness of mental health problems. Any cost saving is likely to be found in reduced admission and transport costs but may not be evident until years into the service.
The prescription data showed pre-existing patterns (high anxiolytics, low antidepressants) in keeping with other research about psychotropic prescribing in general practice. A review of the National Ambulatory Medical Care Survey database showed that general practitioners prescribed anxiolytics most frequently, while psychiatrists prescribed antidepressants more commonly [11]. There again may be a lag in any change in prescribing habits as a result of improved mental health skills among general practitioners.
Many visiting staff commented on the local problems within the area health district organisation. In some areas, it included poor and ambivalent communication between the community health staff and the local general practitioners and in others, a disagreement between general practitioners in a given town. Visiting staff had anticipated that there maybe a problem with intermittent telephone contact seeking assistance about psychiatric matters but in fact this was used judiciously and extremely successfully. Metropolitan visiting staff were impressed with the resilience of their colleagues and the way that problems were largely contained in the local area before extensive support was sought from outside the area.
Most staff have made a further 18 months' commitment to the project. Many of the local health services have made enquiries about the longevity of the project, keen to see the visiting consultative and educative service continue. There have also been enquiries about re-running the education sessions or further teaching material covering new areas. Following completion of the 10 sessions the tutor, participant and evaluation material was collated as a stand-alone resource.
It is proposed to continue the provision of psychiatric clinical and educative services in rural and remote locations in NSW. At present bimonthly visits, providing clinical services every visit, with a mental health skills development workshop every second visit, are being undertaken. These services are being dovetailed with the expansion of Telepsychiatry across the State.
Footnotes
Acknowledgements
This work was kindly funded by Commonwealth Department of Human Services and Health, Rural Health, Support, Education and Training Grant no. 249. The Pharmaceutical Benefits Scheme kindly provided prescription data.