Abstract
The Second National Mental Health Plan [1] has as one of its key themes, a focus on the quality and effectiveness of services with an emphasis on outcomes for consumers and carers. This focus on quality and outcomes results in an increasing interest in clinical practice guidelines, benchmarking of services and the development and evaluation of models of best practice. One way to progress these issues within clinical services is to develop clinical pathways for groups of patients with similar conditions.
Clinical pathways have been defined as ‘clinical management tools that organise, sequence and time the major interventions of professionals for a particular case type, subset or condition’ [2]. Simplistically, a timeline is plotted on one axis with interventions on the other.
The proposed components of service delivery define multi-disciplinary team members' individual responsibilities with expected outcomes. At each stage any deviation from the desired pathway is documented. Such deviations are termed variances and can be attributed to patient, staff or systemic factors. The clinical pathways model allows key decision points to be identified and monitored, and puts treatment decisions closer to the key clinical events [3,4]. Anticipated benefits of the clinical pathway model include improved health care quality by reduction of unnecessary or unhelpful variances and a reduction in unnecessary health care expenditure by reducing duplication and time wasted.
Hopefully this process leads to improved clinical outcomes for patients.
In order to trial the usefulness of the concept, one public sector mental health service elected to develop and pilot the implementation of a clinical pathway for the treatment of major depression in the inpatient psychiatric unit during 1997/8.
ORGANISATIONAL CONTEXT
The Division of Psychiatry and Mental Health is part of the St George Hospital and Community Health Service, and serves a population of 220 000 in the South Eastern Sydney metropolitan area of New South Wales. The Division comprises an eighteen-bed inpatient unit, integrated with two community treatment teams and a twenty four-hour crisis team.
The service already had experience in the use of clinical practice guidelines for the treatment of first onset psychosis [5]. Because these were used principally in the community setting, it was agreed that the piloting of a clinical pathway would occur in the inpatient setting, thus broadening the number of staff in the service who were being exposed to new experiences in clinical practice change and quality improvement. The diagnostic group of major depression was chosen, as it met the criteria of a high volume, high risk and high cost category, as outlined in traditional models of clinical pathway development [3].
METHOD
The aim of the project was to develop and pilot a clinical pathway and to evaluate its acceptability and usefulness to staff.
Engagement of all relevant staff
In March 1997, a working party was convened to develop a consensus clinical pathway for major depression relevant to the St George inpatient context. The working party comprised divisional manager, consultant psychiatrist, nursing unit manager, social worker, clinical nurse specialist, occupational therapist and psychologist. It identified the individual components of comprehensive inpatient treatment of major depression, and included elements of the existing model of clinical practice in the service as well as the elements expected within a broadly agreed concept of best practice which was derived from the consensus opinions of the group.
The initial draft pathway
The individual components were formulated into a written proforma clinical pathway that proposed specific interventions, to be completed by individual professionals, at a specified time. As the service was concerned about average length of stay for patients with major depression, the clinical pathway imposed a suggested cap of 10 days. Documentation of clinical variance was also a key component of pathway development. The first draft of the document was disseminated across a wider group of staff for comment. In its final version, it comprised a 4-page document with 111 activity and variance sections. It was colour coded by professional group to clearly identify the responsibilities of each person at each stage.
Pilot implementation
Pilot implementation commenced in October 1997 and ran over 8 months. Initiation commenced with a series of education sessions for all staff working in the inpatient unit. Thereafter, all patients admitted with a diagnosis of major depression were flagged at the ward's daily clinical meeting. The nursing unit manager then regularly reminded the staff to update the form, with each patient's primary nurse being the designated ‘owner’ of the pathway and ultimately responsible for its completion by the treating team.
Evaluation
Evaluation focused on the acceptance and completion rate of the pathway, rather than on patient related outcome measures. Focus groups were held to establish the subjective views of staff about the design and use of the pathway. These occurred approximately two months after the pilot phase had concluded. The focus groups, combined with variance analysis, aimed to direct future restructuring of the body of the pathway.
RESULTS
Within the first 8 months, 17 clinical pathways had been completed. This represented 6.9% of the total admissions for that period and 58.6% of the total number of patients to whom the pathway potentially applied. The pathways were analysed to identify the number of clinical activities that had been completed and the number of variances recorded. The completion of the pathway showed considerable variation between the professional disciplines. The nursing, social work and medical staff activity sections had been completed most fully while there was a lack of completion of occupational therapy, psychology and case management activity sections. The generic activity sections entitled ‘all disciplines to complete’ were also poorly completed (Table 1).
The activity section completion rate declined during the course of the 10-day pathway but this was not matched by an increase in the completion rate of the variance section (Table 2). No patients were admitted and discharged within the suggested 10-day cap on length of inpatient stay.
Completion rate of the activity and variance sections of the Clinical Pathway by professional group
Completion rate of activity section for each of the 10 days of the pathway
There was a lack of consistency in the type of information entered into each section of the pathway across all professional categories, e.g. variance was often documented incorrectly in the activity section. Some of the pathways were completed by a single professional, most frequently the primary nurse for the individual patient, implying that the responsibility for completion of some of the pathways had been taken by a single person rather than the multi-disciplinary team.
Staff subjective views
From the staff focus groups, feedback from those involved in the development of the project was generally positive. The multidisciplinary nature of the working party and the ability of each professional group to determine their own tasks were appreciated. Some members of the working party had found the process of development challenging, specifically in attempting to fit a global model of mental health care into a specific timeframe.
Feedback from the clinical staff who actually used the pathway was less positive. Many perceived the pathway to be simply a research tool and were unclear as to the clinical benefits in the ‘real world’ of clinical practice. A recurring theme was that there had been inadequate explanation about the use of the pathway and that its implementation had come from the ‘top down’. They felt that the pathway did not help in guiding their daily activities, and reported completing it retrospectively, rather than using it as a tool to prompt their management plans. Nursing staff particularly perceived the pathway to be a ‘burden’ and commented that it was repetitious and had too strong a focus on medication and side effects. Of all the professional groups, the nursing staff, despite having the highest completion rates, felt that the pathway was least useful to them.
All professional groups identified the timeframe of 10 days within which to achieve all the required clinical interventions for treatment of depression as too ambitious in the current clinical setting.
The staff were asked to suggest improvements and revisions which would be required for the pathway to be more acceptable. Suggestions included: incorporating the pathway routinely into the multidisciplinary ward round process; modifying the social work, occupational therapy and psychology requirements to an ‘as needed’ basis; changing the timeline to provide guidelines rather than a fixed pathway. They agreed that the pathway should continue within a multidisciplinary framework, rather than becoming a pathway for nursing and medical staff alone, in order to encourage greater levels of multi-disciplinary team integration.
DISCUSSION
The results obtained from this pilot phase of the Clinical Pathway project have confirmed the views reported within the literature on the difficulty of implementing pathways in routine clinical practice settings.
The literature clearly highlights the need for assertive dissemination strategies [6], coupled with the provision of incentives to complete guidelines [7]. In the absence of either of these, it is positive that within this project, as many as 17 pathways were partially completed. However, the lack of specific training may have contributed to the poor completion rates.
It is noted that the quality of data entered into the pathway proforma was less than optimal and did not always address the desired issues. In analysing the data, we concentrated solely on completion rates, regardless of the quality of information or validity of entry. Such poor quality data means there are difficulties in drawing conclusions about specifics of the pathway other than its completion rates. Since the success of any pathway is related to its effect upon patient outcomes, poor quality of their use will impede any significant impact on outcomes [8].
The variation between professional groups is of some interest as it reflects both differing levels of interest in the project and wider organisational structure issues. Those interventions under the control of senior medical staff would generally be expected to receive more supervision and this could explain higher medical completion rates. The commitment of the nursing unit manager would account for the higher level of nursing staff input. The low numbers of psychologists and occupational therapists employed within the inpatient unit at the time of the project is a systemic explanation for their low uptake rates.
The lack of involvement of case managers is not so easily explained. It is possible that their main activities, as determined by the pathway, related to discharge planning, and therefore the limit of 10 days prevented them from complying within the timeline. It is more likely that they perceived the inpatient pathway as irrelevant to them as they had not been involved in its development, and this is a significant flaw within the original methodology. Any attempt to develop a pathway for use by a particular staff group must involve that staff group from the outset.
It should be noted that many of the staff participating in the focus group review process had not been involved in the original development and implementation of the pathway due to staff turnover. This would explain a perception that the original educational packages were inadequate and underscore the general lack of enthusiasm for the pathway.
In keeping with the Second National Mental Health Plan [1], there is a desire for consumers to become involved in the development of clinical pathways. Consumer participation was lacking in the original working party and this omission would need to be addressed in future planning.
This pilot study was implemented within the in-patient unit. As described above, even within this relatively simple model, difficulties were encountered. The ideal ultimate model of integration of the pathway across the whole service, incorporating entire episodes of care, is likely to increase the complexity of all these methodological difficulties.
CONCLUSION
Our pilot study has provided us with a greater understanding of the difficulties of designing and implementing a clinical pathway in the public sector mental health services. The project introduced the concept of systematic and timely management of care to clinicians, but it is clear that there needs to be further development of the implementation component if it is to become a routine part of local practice. In addition, components of the pathway need redefining, with some restructuring of the pathway and variance sections to meet the needs of managers and clinicians.
The challenges for those senior managers and clinicians who wish to drive this process forward are:
To develop a model of education and dissemination of the pathway to ensure a higher rate of completion and uptake. Suggested strategies would include identification and targeting of each professional group for education [9] and the use of specific educational packages rather than general feedback or information [4].
To provide incentives to all members of the multidisciplinary team to complete their components of the pathway. Since the use of incentives at all levels within the public sector of health is poorly understood this may be a major issue.
To ensure adequate matching of available staff resources with the requirements of the pathway.
To develop a more sophisticated pathway initially to span the entire inpatient stay, with realistic time limits on lengths of stay targets, and ultimately to span the entire episode of care including community mental health interventions and general practitioner's involvement.
To develop and implement a pathway that incorporates measures of clinical outcome. Our pilot study focused principally on an evaluation of the completion rate rather than patient-related outcomes. In developing our pathway to its next stage, we aim to incorporate a measure of outcome, e.g. an illness severity rating scale at entry and exit, such as HoNOS [10].
To develop a learning organisation, wherein clinicians improve their receptivity to new practices, and new practices become self-sustainable.
In conclusion, our initial experience with implementation of clinical pathways has convinced us that it will be useful to continue with the concept despite the more intensive input that will be required. Our review of the literature indicates that clinical pathways represent a useful model for mental health services, particularly in the area of quality improvement. Future developments will need to address the implementation of pathways, evaluation of outcome measures and the incorporation of clinical practice guidelines. One suggestion is to develop national guidelines for the development of clinical pathways in a mental health setting, similar to the NHMRC ‘Guidelines for Guidelines’ [11].