A feasibility study of expert patient and community mental health team led bipolar psychoeducation groups: implementing an evidence based practice

Bipolar disorder is one of the largest medical contributors to disability [1]. It is a highly recurrent disorder with a lifetime prevalence of around two percent. Approximately 10 percent of patients will end their life by suicide [2]. Bipolar disorder is associated with interpersonal stress, job related difficulties, high unemployment, high dependence on benefits, low annual income, high work absenteeism, low work productivity, poor overall functioning, lower quality of life, difficulty maintaining long-term relationships and reduced overall social, physical and emotional well-being [3].

Relapse rates are reduced, but not eliminated, in efficacy trials of mood stabilising drugs such as lithium and sodium valproate. However, there is a significant efficacy-effectiveness gap and response is often associated with persistent neurocognitive impairment [4] and psychosocial dysfunction.

Psychoeducation is widely recommended [5, 6] and has randomised controlled trial evidence for an increase in time to manic [7] and depressive [8‐10] relapse in the short term. In addition Colom’s group [11] showed significant reductions at 5 year follow up in number of illness episodes, days spent unwell, hospitalisations per patient, and days in hospital per hospitalisation. Despite this, psychoeducation is not widely available in routine care, partly because its delivery is not supported by current service models. It therefore appears prudent to seek to implement this evidence based practice [12].

Community mental health teams (CMHTs) are the standard service providing input for patients with secondary level needs in the UK. They are multidisciplinary teams of medical, nursing, social work, occupational therapy and/or clinical psychology staff usually in a community base away from the hospital, which provide assessment and treatment to those with enduring mental health conditions. However the evidence that standard CMHT input helps patients with bipolar disorder is insubstantial [13].

There is an increasing use of peer specialists (PS, defined as individuals with severe mental illness who use their experience to provide services for other people with severe mental illness) in mental health services, particularly in the United States of America [14, 15]. It is argued that this development is supported by social learning theory [16] and that PSs can effectively teach coping and self-management strategies [17] and further that PSs often increase patients’ engagement and improve patient satisfaction [14, 18]. There is now experience of PSs facilitating groups in mental health settings e.g. [19].

Our model proposes a manualised psychoeducation group delivered by CMHT workers and PSs working together under expert supervision within the CMHT. These groups should be financially more viable than groups provided by a specialist mood disorder service [11], or psychologists. Moreover, the groups should help enhance the skillset of CMHT workers and should allow the intervention to be delivered by the service which will also be providing other aspects of care including crisis response.

This paper outlines a feasibility study primarily run to assess the acceptability of the proposed group psychoeducation programme and to identify drivers and barriers that may impact on the rolling out of the programme. It outlines how these were identified and addressed, and examines whether groups can be successfully run in this format.

The pilot research was conducted in accordance with the Helsinki Declaration and prior approval of the Local Research Ethics Committee was obtained.

19 trainee facilitators attended the initial three day training. There developed a core group of 8–10 trainee facilitators who then attended the twelve weekly sessions that followed. Eight trainee facilitators filled in the semi-structured questionnaires at the completion of the training. Eight trainee facilitators views were gained via discussion, regarding the training process.

The study provides evidence that the delivery of psychoeducation groups using community mental health team staff and PSs is acceptable to participants and facilitators if they are well supported. It illustrates the required organisational steps required to successfully run such groups. It also illustrates the value of using a variety of qualitative and quantitative research methods to systematically elicit barriers and drivers to implementation within an organisation learning framework so that lessons learnt can be readily applied to improve the implementation of the intervention.

There are a number of methodological limitations to the approach we took. The first is consideration of the method of knowledge mobilisation and implementation research. There are many approaches to both and many of these approaches are reviewed by Rowley and colleagues [12] and Ferlie and colleaguesl [35]. While there is a more settled literature on implementation of evidence into clinical practice in general, there is much less agreement on the best way to do this in specific health care organisations, as we describe here. Therefore, investigators have been invited to consider the theoretical and method approach that is most suitable to the question that is being studied [35]. Organisational learning is an approach that helps business to thrive in changing market conditions but it has rarely been evaluated in health care research and so its validity and utility in health service practice is yet to be established [12]. It values competency in delivering interventions and takes the perspective that a systematic understanding of an organisation’s cognitive capacity at each level of the organisation and as a whole is critical to the delivery of such interventions [35]. To be effective in organisational learning, research methods have to be analysed in time to inform further refinement of interventions while there is still enough momentum and leadership within the health care organisation to deliver them. The data and changes to address the barriers to the delivery of the intervention are an iterative process. The emphasis is on understanding the sociological context of the changes, especially absorptive capacity or the ability to utilise new knowledge at that time at different levels of an organisation [36]. The iterative nature of the implementation approach shares many similarities with participatory action research or continuous quality improvement [37]. Compared with longer term systematic plans of implementation based on a readiness for change model e.g. [38] or other projects used to implement service change in serious mental illness e.g. [15] it is more reactive and concerned with what can be done now. The systematic collection of data to consider barriers at organisational levels and user experience makes this a broader and more dynamic process than could be achieved merely through professional reflective practice.

Potentially there are many limitations of the approach. There may be insufficient time and resource to adequately explore discrepant data that might be important to fully understand barriers to the delivery of an intervention. For instance, in this study we were unable to interview participants who dropped out of the groups who may have been able to pinpoint further barriers to the intervention and probably did not have the same satisfaction with the intervention. We did not interview other clinicians and administrators involved in the care of the patients in these groups or who might refer patients to such groups due to the limitations of the pace of the work and resource to do it. The analysis was also performed by researchers from only one professional background (psychiatry) and might have been enhanced by consideration of the data from a variety of different professionals and service users.

To improve the internal validity of the research methods, we utilised triangulation of research methods and data sources (direct observation, participant perspectives, facilitator perspectives) [21]. However, there is no common agreement about the optimal way to use this technique [22]. There can be incompatibilities between the units of analysis and the research paradigms that might contradict each other or even amplify sources of error and bias [25, 39]. In this instance, the methods can all be seen as inductive and focussed on eliciting barriers and enablers to a specific intervention so the most likely error if present at all is to underestimate or to fail to identify some of the barriers to the intervention. The organisational learning process as described here seems to have been effective in addressing the problems that were within the control of the research team. However the barriers to the delivery of the intervention within the health care organisation itself were not addressed sufficiently as none of the research team could provide the necessary senior clinical leadership within the health care organisation that hosted the study to address these barriers [40]. Furthermore there were barriers to the use of the crisis plans developed by participants in the groups that were not enacted upon by primary care, accident and emergency and crisis resolution and home treatment teams. The lack of knowledge concerning how to utilise these plans might be addressed through education. This would require leadership and planning outside the mental health care organisation as well as within; an argument for a more planned approach to implementation [15].

These groups represent a different model, allowing health professionals and service users to train and work together. The results were particularly positive with regard to participant acceptability. There were high satisfaction ratings, good self-reported goal attainment, and good participant retention. As we had hoped PSs and health professionals did work well together within the treatment groups to create an excellent environment in which patient learning was optimised. This project set out to be collaboration between patients and professionals and this may explain, in part, the high level of acceptability. The experiences of the PSs in this programme mirror the experiences of PSs in other programmes [41]. Ideally there would also be further work on making training more flexible for PSs, developing clear guidelines on readiness to start training to facilitate, and finding flexible ways to reward PSs.

The major issue for on-going implementation for this programme was the inability of the project team to gain enough senior management support to secure longer term funding for key posts. The inability to provide a stable long-term framework for the groups, felt particularly unsatisfactory, as we became increasingly aware of the significant personal investment many PSs were offering. Although there is some evidence that savings in in-patient and other emergency care more than paid for the intervention over a five year period [42], there are undoubtedly initial costs to the programme, in this case nurse specialist and some consultant psychiatrist time. In order for the programme to be sustainable this initial cost would also need to include “backfill” time for the professional facilitators. Importantly, this study took place in the context of a British national health service which was, and is, trying to implement cost savings.

In terms of limitations this pilot has not examined efficacy, but aspects of the intervention’s efficacy have already been demonstrated [7‐11]. This study has informed the design of another group psychoeducation intervention involving health professional and expert patients by one of the authors (RM) [43]. However, the new study is not being performed in the same health care organisation as the current study and is not involving local community mental health team staff in delivering the intervention. As a result compared to the current study, the implementation of the intervention into routine clinical practice might be faced with an additional barrier to overcome, namely its acceptance into the service delivered by local community health team.

We conducted this pilot in the UK and have considered it from a UK perspective but recognise that the advantages of making psychoeducation widely available are not confined to our shores and that the ability to deliver such an evidence based intervention to patients with bipolar disorder by training and developing peer specialists and by creating a supportive organisational structure is a global issue.

Our experience suggests that it is feasible to run a group psychoeducation programme for bipolar patients, with health professionals and PSs working together as facilitators to provide high levels of acceptability for group members. Patients working as facilitators and shaping the intervention may have been important in achieving these high satisfaction levels. The intervention was not implemented into routine practice. However, the feasibility study has allowed us to highlight barriers that would need to be overcome to allow the successful expansion of this programme. These included: maintaining structures for facilitator support; finding ways to allow patients to work sustainably within NHS structures; training staff involved in the delivery of crisis care and on-going support to utilise the early warning signs intervention; obtaining advocacy and support from senior clinical or managerial leaders in the health care organisation to secure long-term funding for a bipolar disorder nurse specialist; and the release from other duties for other health care facilitators.

KC is a senior psychiatry trainee, DP held the position of bipolar nurse specialist and is now ward manager on an inpatient unit. CB was a community psychiatric nurse but is now Senior Lecturer. SW is a Senior Lecturer and honorary consultant psychiatrist. RM is the Principal Investigator for PARADES Group psychoeducation versus Group Support RCT, Chair of the NICE Guideline Development Group for Bipolar Disorder (revision) and Director of Research for NIHR CLAHRC NDL which is researching organisational learning as a method of improving implementation. During the writing up of the project, RM was Director of Research for the National Institute of Health Research (NIHR) funded Collaboration for Leadership in Applied Health Research and Care for Nottinghamshire, Derbyshire and Lincolnshire (CLAHRC-NDL), funded by the National Institute for Health Research Collaboration for Leadership in Applied Health Research and Care for Nottinghamshire, Derbyshire and Lincolnshire (NIHR CLAHRC NDL). The views expressed in this manuscript are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health.