A Longitudinal, Comparative Evaluation of Reflective Practice Groups for Nurses Working in Intensive Care and Oncology

This paper presents a longitudinal evaluation of Reflective Practice Groups (RPG) for nurses from Intensive Care and Oncology settings at Nambour Hospital, Queensland. The literature review examines previous evaluations of RPG, establishing the contribution of the current study as its longitudinal, quantitative comparison of RPG from two distinct nursing contexts. The two year evaluation utilised a validated tool, the Clinical Supervision Evaluation Questionnaire (CSEQ), designed specifically for facilitative group supervision. The results support the positive findings of a pilot study by the author, with a majority of nurses from both groups rating their experience of RPG as ‘definitely positive’ throughout the evaluation period. Subtle differences in data from CSEQ subscales of purpose, process and impact indicate that aspects of, and benefits from, RPG were experienced differently in each group. Whilst both groups reported positive perceptions of process, the ICU group showed a notable positive shift in fields relating to trust, respect and safety whereas data from the Oncology group indicated an upwards trend in ratings of impact fields such as self awareness, clinical insight and quality of care. These results are discussed in relation to group characteristics and workplace contexts. Despite limitations, the data is felt to reinforce the value this RPG model places on facilitation techniques and management of group process as means of enhancing critical reflection, promoting a focus on the interpersonal aspects of care giving and encouraging supportive interaction. The study establishes a sound basis for ongoing research into this RPG model. *Corresponding author: Chris Dawber, Researcher, Consultation Liaison Psychiatry, Nambour Hospital, School of Medicine, University of Queensland, Queensland, Australia, Tel: +418227726; E-mail: Chris_Dawber@health.qld.gov.au Received October 04, 2013; Accepted November 28, 2013; Published December 05, 2013 Citation: Dawber C, O’Brien T (2013) A Longitudinal, Comparative Evaluation of Reflective Practice Groups for Nurses Working in Intensive Care and Oncology. J Nurs Care 3: 138. doi:10.4172/2167-1168.1000138 Copyright: © 2013 Dawber C, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

For the past seven years the Researching author has been facilitating Reflective Practice Groups (RPG) for general nurses working in clinical settings. RPG provide a space for nurses to explore interpersonal aspects of their work [1]; difficult and distressing clinical situations, workplace stress, emotional and behavioural issues with patients, aspects of counter transference, existential and ethical dilemmas [2]. As such, they can be viewed as a form of peer group Clinical Supervision (CS).
Previous papers by the author outline the development and evaluation of a particular RPG model designed specifically for nurses working in clinical settings [3,4]. This current study explores the practicalities of implementing nursing RPG further and examines their impact in two very different nursing environments. The study notes changes in nurses' perceptions of the purpose, process and impact of RPG over a two year period; reprising the quantitative element of the original pilot study [4]. The RPG in the study were drawn from a cohort of 15 groups currently being facilitated at Nambour Hospital in Queensland, Australia. The Intensive Care Unit (ICU) and Cancer Care Day Services (Oncology) represent quite different nursing contexts and, whilst the underpinning model for RPG remains the same for both units, these differing clinical contexts are felt to influence some aspects of the ways each group operates.

The Group Model
The RPG model evaluated by this study evolved over a period of several years [3]. Similar to other forms of group supervision [5][6][7], the model utilises concepts from object relations [8] and self psychology [9]; in particular a focus on 'group-as-a-whole' [10] and 'group self ' [9], to inform a process-focused facilitation framework. 'The group' is viewed as an entity in its own right; with the potential to regulate, restrict or enhance the reflective process. Each session develops from clinical narrative provided by participants and critical reflection [11] is prompted through facilitator interventions directed at 'the group' . Topics cover a range of clinical, professional, ethical and organisational issues; however there is a particular focus on the interpersonal aspects of nursing care.
Issues of staff resistance and workplace culture have been obstacles to the establishment of Clinical Supervision (CS) groups for nurses in the past [12][13][14], however this model utilizes a collaborative approach; actively engaging nurses in the establishment of RPG in their own workplaces [3,15]. Whilst reflection on practice is an important component of CS [16][17][18], CS has historically been viewed with suspicion by nurses; wary of other functions such as professional monitoring and clinical governance [19,20]. Several authors have sought to differentiate types of staff groups [13,21,22]. The Minnesota Association for Children's Mental Health [23] provides a definition that distinguishes reflective supervision from both administrative and clinical supervision by its focus on "shared exploration of parallel process... ", "professional and personal development within one's discipline by attending to the emotional content of the work" and a "greater emphasis on the supervisor/consultant's ability to listen and wait, allowing the supervisee to discover solutions, concepts and perceptions on his/her own without interruption" [23]. Whilst the title RPG has been used to describe a range of groups involving general nurses [18,[24][25][26][27], there does not yet appear to be consensus on how this name relates to group process and function. In this study the title RPG is used to clearly articulate the primary group function as reflection, align the groups with principles similar to those outlined by the MACMH [23] and address resistance by tapping into a growing interest in RP amongst nurses [28]. Regardless of these considerations, it is acknowledged that RPG and CS groups have many similar aims and goals and may be viewed as variations of the same activity.

Research Questions
• Can the positive results from a pilot study [4] be replicated with two new groups of nurses from significantly different clinical contexts?
• What can the data show us about the way that individual participant perceptions of the purpose, process and impact of RPG evolve over a period of time?
• What can the data show us about the ways that collective/group perceptions of the purpose, process and impact of RPG evolve over a period of time?
• What are the differences and similarities in the data from the two groups, and how can this inform the development of the model?

Ethical Issues
Research was approved by the Human Research and Ethics Committees of both The University of Queensland and The Prince Charles Hospital, on behalf of Queensland Health and the research has been cosponsored by these two organisations. The Sight Specific Application was granted by Sunshine Coast Hospital and Health District. Permission to use the Clinical Supervision Evaluation Tool (CSEQ) was provided by Dr Simon Horton, on behalf of its authors [29].
It is the author's standard practice to periodically evaluate RPG with the CSEQ, for quality improvement purposes. All RPG participants are made aware that de-identified data from these evaluations may be used for research purposes. Ethics approvals have authorised use of this retrospective data. Participant information, consent forms and withdrawal of consent forms are provided where additional data is collected specifically for research purposes.

Literature Review
The literature review in the pilot study [3] identified articles on RPG for general nurses in clinical settings published prior to 2009 [22,[30][31][32][33][34][35]. The current review identifies similar articles published since; particularly those utilising quantitative methodology to evaluate RPG. A search of CINAHL and PsychNet databases, using the words Reflective Practice Groups, Nurses and General Nursing, identified 533 articles relating to RPG. 86 of these articles related in some way to Reflective Practice Groups for Nurses with 32 of these published since 2009. 15 articles addressed Reflective Practice Groups in general nursing. Other articles addressing related areas, such as group supervision (GS), were identified through citation.
RP is an important mechanism of professional [36] and personal [37] development for staff working in a range of human service settings [29,33,38]; education [39][40][41], psychology [42], social work [43], other allied health disciplines [29,44], medicine [45,46] and health management [19,47,48]. Most research into RP in nursing occurs in the nurse education sector [1,12,16,39,49] and the majority of articles on nurses in clinical settings address group supervision (GS) for mental health nurses [8,15,[50][51][52]. Many articles are discursive or qualitative in nature and it is generally acknowledged that there is still limited quantitative data measuring the impact of RPG or GS for nurses on individual participants, groups or organisations, let alone patient care [50,53,54].
One early multi-methodology evaluation of RPG for student nurses [55] incorporated a postal questionnaire. Of 72 responses, 83% quantitatively rated RPG as 'useful' or 'very useful' , with the qualitative component supporting contemporary studies by Platzer et al. [27,56] highlighting the importance of facilitation and group process. A more recent study evaluating the experiences of 126 professional and managerial participants from an action-learning group [48], used a purpose produced 31 field questionnaire containing both quantitative and qualitative elements. The quantitative component provided an overall mean rating of 7.4 out of 10 for the groups whilst qualitative feedback indicated that participants 'learnt more during action learning sessions than from them' [48].
Knight et al. [42], report on the development of a new tool for use in a pilot study of RPG involving 124 psychologists. The RPG Questionnaire (RPGQ) is a 113 item tool comprised of 98 Likert scales and 15 qualitative questions. 71% found RPG valuable however 43% also found the groups distressing. As well as some interesting correlations between participant distress and RPG effectiveness, this study provides further insights into participant perceptions of facilitation factors; '58% of respondents (n=64) said they would prefer the facilitator to be more active, more challenging, more experienced, more theoretical, more containing or more dynamic... 15% (n=17) felt they would have liked a clearer rationale/explanation regarding the purpose of the group" [42].
Another recent study [57] outlines the development of a validated 36 item tool, the STERLinG, which assesses facilitator competency and effectiveness in small groups run for medical and allied health trainees. Development of this tool included two pilot applications involving 964 students from a range of Universities. Preliminary results directly link the development of self awareness and reflective thinking with process management. The STERLinG may be relevant to the evaluation of nursing RPG and and its use may be sought in future research by this author.
There are few articles quantitatively evaluating group formats in general nursing settings. A series of 3 questionnaires, adapted from a previously unpublished but reportedly validated tool, evaluated the experience of 24 ward managers participating in a CS group in a general hospital [19]. Findings were positive and similar to those of the author's pilot study [4].
Brunero and Lamont [58] describe a 6 month trial of GS involving 9 groups of nurses from a range of specialty areas at a large tertiary hospital. Evaluation incorporated a 21-item evaluation tool [59]. Results indicated "a strong sense of the CS taking a 'normative' focus, particularly around practice standards and models of care" [58]. Whilst there was overall support for the program within the hospital, criticisms related to consistency and facilitation style. Logistical issues affected the sustainability of the program past the evaluation period [58].
With 'over 90 licensed studies worldwide' [51], the most utilised and validated evaluation tool for nursing CS is the recently revised Manchester Clinical Supervision Scale (MCSS-26). Its authors recommend a number of considerations when implementing and researching supervision. These considerations include selecting a discreet clinical location and involving all staff wherever possible, enlisting managerial support, adequately preparing and supervising facilitators, and employing standardized group protocols. [51].
It would seem logical that similar considerations should guide the implementation and research of RPG. The groups in this study, and the associated evaluation, meet most of these guidelines; however whilst group frequency and length are set for each specific RPG, they differ between groups due to the way the groups are set up to suit specific clinical contexts. Other deviations from these recommended protocols are group size (at times as high as 12) and the fact that, whilst all unit staff are recruited during the establishment phase of RPG, work place issues continue to affect attendance and retention. The 5 day a week, 9am-5pm context of the Oncology group (n=20), for example, allowed regular attendance by a stable core group of staff, whilst the necessities of shift work, higher clinical acuity, larger numbers of nurses (n=100-120) affected attendance and retention of participants in the ICU group.

Methodology The evaluation tool
The Clinical Supervision Evaluation Questionnaire (CSEQ) is a 14 item validated evaluation tool developed by Horton et al [29]. to evaluate group supervision for speech therapists. It has been found to have reliability and validity in its original context and potential utility for CS groups with other health professionals.
Whilst the MCSS-26 remains the evaluation tool of choice for mental health nursing supervision [51], the CSEQ was specifically designed to evaluate group supervision that utilises a facilitative approach to encourage reflection. The CSEQ is also relatively brief and easy to complete, taking only 3-5 minutes; an important consideration for nurses working in busy clinical environments.
The 14 items of the CSEQ are divided into 3 subscales addressing participant perceptions of group purpose (3 questions), process (5 questions) and impact (6 questions). Respondents grade each field using a 5 point Likert scale. The lowest possible score is -28, indicating strong disagreement with all fields, and the highest possible score is +28. A score of +14 represents a "definitely positive view of the groups" [29] and indicates an average mean score of +1 for each field. Any score of less than 0 indicates a negative view of the group.
The CSEQ format used in this study was slightly modified with the permission of its authors. The name of the activity being evaluated was changed to 'Reflective Practice Group' and several demographic fields were added to allow comparative review: 'Pseudonym' , 'Ward' and 'Number of groups attended' . A preamble provides participants with rationale and guidelines for the questionnaire. Three additional questions addressing the continuation of RPG have been appended outside the parameters of the CSEQ, and following an official acknowledgement of the authors. None of the changes are felt to have substantially altered any of the 14 data collection fields.

Participants
The Nambour Chemotherapy unit operates Monday to Friday and is staffed by a multidisciplinary team including 20 nurses. Cancer care nurses often develop strong bonds with their patients; working with them for significant lengths of time and "bearing witness" [60] to their journey as they progress through treatment to either remission or, in some cases, death. Oncology nurses define their specialty as one that requires compassion and acknowledge the interpersonal aspects of patient care as key elements of their practice.
The Cancer Care Service was the first area to commence RPG at Nambour Hospital. The concept had previously been introduced by a clinical coach and several RPG, addressing technical aspects of the job, had already occurred. Collaboration between the author, the clinical coach and nursing staff resulted in RPG evolving to incorporate a stronger focus on interpersonal aspects of working with cancer patients. The group also addresses organisational and interpersonal issues affecting team cohesion. It runs monthly, for an hour; utilising a dedicated morning education timeslot. It is well attended, with membership at any given group varying between 6 and 12. Group membership is almost exclusively female. Participants have anywhere between 1 and 30+ years of nursing experience. It is a semi-closed group for nurses on the unit; however students and graduate nurses on placement are invited to attend.
Nambour ICU is a 12 bed unit with proportionately high numbers of both medical and nursing staff. During the evaluation period, nursing staff numbers on the unit fluctuated between 100 and 120: approximately 15% being male. ICU nurses deal with significant trauma and serious illness and there is a strong focus on technical aspects of the job. Nurse patient ratios are at least 1:1, although many patients are unconscious or delirious for a large proportion of their admission. ICU nurses develop a unique relationship with these totally dependent patients; anticipating their needs, providing comfort, care and at times advocacy. Approximately 13% of ICU admissions relate to drug overdose [61] and ICU nurses regularly deal with the emotional distress and disturbed behaviour of these patients as they recover. They also provide support and reassurance to distressed and anxious relatives.
The ICU RPG commenced several months after the Oncology group. Whilst the establishment of the group was an initiative of the CL nurse, it involved collaboration with management, nurses on the unit and ICU nurse education staff. The group primarily addresses confronting clinical incidents and the management of challenging patient behaviours. At times it may also be used as a forum to discuss contextual, organisational or ethical issues. ICU RPG runs fortnightly, in the afternoon, for staff coming on duty. Attendance is usually 6 -12 nurses; often including students. It lasts for 45 minutes and timing is important as any delay impacts on the outgoing shift being able to leave. The group is occasionally cancelled due to acuity on the Unit or mandatory training requirements.

Results
The Oncology RPG completed 5 evaluation rounds using the CSEQ, with the ICU Group completing 6 ( Table 1). 58 questionnaires were returned by 30 participants from Oncology; all fully completed and all with pseudonym identifiers. 76 questionnaires were returned by 58 participants from ICU. 8 of these were only partially completed and 12 were submitted without a pseudonym identifier; making it impossible to determine whether these were first time or repeat responders. All questionnaires were included in some aspect of the study; with partially completed questionnaires contributing data to analysis of specific fields but not total scores and those without pseudonyms being included in collective group scores for specific rounds.
Data was collated using Excel spreadsheet and then analysed by the researcher. Correlations were made by filtering the data from relevant fields. Data correlations were rechecked and reviewed a number of times before a summary of the collated data was submitted to the research supervisor for further review and comment. Table 1 provides an overview of the collated data from all evaluation rounds for both groups. In keeping with the CSEQ author's guide to interpreting results, and consistent with the group framework adopted by the RPG model, a collective mean group score of 14+ is seen to indicate that 'the group' has a definitely positive perception of RPG.
Only 40% of the Oncology group gave a rating of 14+ (definitely positive) in round 1, correlating with lower ratings in the purpose and impact fields of the questionnaire. Subsequent rounds consistently show 75% or more participants rating their overall perception of RPG as being definitely positive.
In comparison, at least 50% of the ICU group perceived RPG as definitely positive in each of the 6 evaluation rounds. The initial ICU evaluation round was affected by a high number of incomplete questionnaires. The lowest percentage of definitely positive ratings for ICU (50%) occurred in round 5, however the mean group score for that round remained definitely positive (14.6). The reason for this anomaly is not clear but does indicate that the 50% of participants rating the group as definitely positive (14+) must have provided scores at the high end of the positive range. (Data from partially completed forms has been used for the specific fields in these tables.The word 'Significant' identifies a particularly positive collective score)  The aggregate percentages for the two year period indicate that well over 50% of participants from both groups perceived RPG as definitely positive, however the aggregate Oncology group percentage was higher (74% compared to 66%). The aggregate mean scores for the evaluation period were comparable (15.8 for Oncology and 15.4 for ICU).
There were just two negative evaluations (-0) over the two year period. Both were from Oncology RPG; Round 2 (score -1) and Round 5 (score -4). The Round 2 negative score was provided by a participant who subsequently submitted two further evaluations; in Round 3, this participant rated +3 and in round 4, +6, indicating an overall positive trend in perceptions of the group. The negative response in Round 5 was from a novice participant (1-2 sessions) who did not submit another evaluation. Table 2 compares mean scores from first and last evaluation rounds for each group; indicating where group perceptions have changed. A score of +1 in a field indicates that 'the group' agrees with the statement defining this field and a rating of +2 indicates 'strongly agree' . The word SIGNIFICANT has been used to highlight mean scores exceeding +1 (i.e. more than agree).
Purpose Fields: Field 1 indicates that both groups maintain a clear perception that "the purpose of Reflective Practice Group is to improve client care". Field 9 relates to the purpose of RPG being to enable clinician confidence and shows a significant increase for the Oncology group. Field 14 indicates that both groups have developed a clearer understanding of what they want to get from attending RPG. Process Fields; 2, 3 and 5, indicate that perceptions of safety, respect and trust within each group have either remained strong (Oncology) or increased over time (ICU). These fields have particular importance in view of the emphasis the model puts on facilitation of group process. Field 10 shows a modest increase for both groups in relation to the confidence individuals felt, bringing issues to the group. Field 13 shows the ICU group maintaining that "there are well established ground rules in my group", whilst the Oncology group's perception of group rules has strengthened. Impact Fields; Field 4, "I have gained new clinical insights through Reflective Practice Group", shows a marked increase in mean scores for the Oncology group whilst the ICU data reveals relatively strong ratings over time. Field 6 and 7 relate to quality of care and self awareness; once again showing significant increases in mean scores for the Oncology group. Field 8, "Reflective Practice Group has helped me feel more confident about dealing with my job", shows increases in both groups, as does field 11; "Reflective practice Group has helped me cope with any stresses at work I may have". The final impact field, 12, relates to the formative field "areas of skill I need to improve". This field shows higher ratings, but an overall decrease for the ICU group, with more modest ratings but an overall increase for Oncology. Table 3 shows collapsed (overall) mean scores for all 3 subscales.
From this data we observe a positive shift in overall mean scores for both groups in each subscale. The most significant change for the ICU group is in the perception of process, whilst the Oncology group shows significantly higher overall scores for both purpose and impact. These graphs show that the data from each of the groups consistently indicates a strong perception of group process (+1). The Oncology group's perception of the impact of RPG fluctuates but trends upwards over time. ICU perceptions of purpose and impact appear reasonably constant over 6 rounds of evaluation (especially latter rounds). There were 12 Oncology respondents who attended 12 or more groups but 0 ICU respondents. It is unfortunate that a significant number of respondents (n=32) did not state the number of groups attended. 100% of participants who attended 7-12 groups rated RPG as definitely positive (14+). 75% of the Oncology participants who attended 12+ RPG rated them as definitely positive.    11 Oncology participants completed serial CSEQ; 5 completing more than 4 evaluations in a row. 6 of the serial questionnaires indicated a trend of increasing scores over the evaluation period. 4 serial scores fluctuated but remained definitely positive (+14). 1 participant gave overall scores that decreased over two rounds (Round 2 = 15 & Round 3 = 10). 5 ICU participants completed 2 serial CSEQ but none did more than this. All of these showed improved scores in the follow up evaluation round.

Discussion
The data from this evaluation supports the positive findings of the pilot study [4]. The aggregate percentage of 'definitely positive' responses for the ICU group (66%) matches the 67% aggregate from the pilot, whilst the Oncology group results show an even higher aggregate percentage of positive responses (74%). The Oncology group had a smaller pool of nurses (20), more consistent attendance rates and proportionately higher numbers of completed evaluations per round.
Areas of particular interest to the author concern participant perceptions of process and impact. Oncology data indicates an increase in participant perceptions of the group's impact over the two year period; particularly areas of self awareness, clinical insight and quality of care. This may be seen to indicate a formative function [62] of RPG.
Interestingly, the ICU group; with its larger pool of nurses (100/120), less frequent individual attendance rates and proportionately less CSEQ returns, still returned a similarly positive overall group mean evaluation score (15.4 compared to 15.8 from Oncology). The most marked change in collective participant perceptions for the ICU group was in the process subscale (particularly areas relating to safety, respect and trust). This could be viewed in relation to developing cohesion within the group, and possibly within the broader work place, indicating a possible restorative effect of attending RPG [62].
Of course there are a range of contextual variables that need to be considered alongside the data; amongst them the impact of factors such as staff turnover, changes in work environment, work group and participant specific issues. Having noted this, the fields on the CSEQ do prompt participants to relate responses specifically to their experiences in RPG and, along with the relative consistency of positive data over the two years, this helps strengthen the connection between the data and participation in RPG.
One interpretation of the data is that a focus on facilitation of process has benefitted both RPG, but in somewhat different ways. Participation in RPG appears to have stimulated reflection (arguably more so in the Oncology group), but has also promoted cohesion and supportiveness within each group. This model for RPG provides an opportunity for, and mechanisms to encourage, reflection rather than a rigid framework or prescribed content. The facilitator collaborates with 'the group' to address a range of additional (explicitly and sometimes implicitly) negotiated tasks relating to process. These tasks need to be worked through in order to maintain group homeostasis. Reciprocally, this homeostasis is felt necessary to enable for 'the group' to move towards deeper and more challenging levels of reflection. Whilst data from individual evaluations show a degree of variation, consistently positive collective mean scores can be seen to indicate equivalent overall benefits from attending either 'group' .
The main limitations of the study relate to methodology. The author works in a busy clinical service and it has been a struggle to fit research in with other commitments. For this reason, some evaluation rounds have smaller samples, or have been collected over a period of weeks. In addition, incomplete CSEQ returns and the omission of coded identifiers by some participants have adversely affected some comparative aspects of the study. Another limitation, and one that is shared by many studies of group CS, is the inability to draw a connection between participant perceptions of RPG and any actual impact on patient care [52].
The CSEQ, whilst efficient and effective for evaluating RPG, does not have the supporting evidence or widespread utilisation of the MCSS-26 [51]. This makes it difficult to compare the evaluated model against other forms of RPG and/or GS. Whilst the author plans to continue using the CSEQ for ongoing longitudinal evaluations and quality improvement purposes, the aim is to seek permission to use both the MCSS-26 and the STERLinG [57], in further research. It is also envisaged that future research will involve greater numbers of participants, multiple RPG from more diverse clinical contexts and different facilitators using the same facilitation model.
Despite the limitations noted above, this study provides further evidence to support the effectiveness of RPG. It is hoped that future research; undertaken in collaboration with more experienced researchers, utilizing more refined methodology and improved data analysis techniques, will build on the evidence from this study to show that the effective facilitation of RPG for nurses, using this model, enhances critical reflection, provides a valuable and effective means of support for nurses, builds cohesion and effective communication in the workplace, and allows nurses to process issues related to the interpersonal aspects of care giving.