Staff experiences related to implementation of a recovery-oriented nursing programme in psychiatric inpatient care
Authorship statement: All authors made substantial contributions to conception and design, or data collection, or analysis and interpretation were involved in drafting the manuscript or revisiting it critically and gave final approval of the version to be submitted. All met the authorship criteria according to the latest guidelines of the International Committee of Medical Journal Editors.
Declaration of conflict of interest: There are no conflicts of interest to be reported for this study.
Abstract
Nursing in psychiatric inpatient care is peripheral to a dominating biomedical model of care. Efforts are being made to implement nursing models based on core values and theories for nursing, such as recovery-oriented practices. The aim of the study was to explore experiences of a recovery-oriented nursing programme (Steps Towards Recovery, STR) among nursing staff in psychiatric inpatient care and their ratings of stress (Maslach Burnout Inventory scores), quality of care (Quality of Psychiatric Care - Inpatient staff scores) and satisfaction with nursing care (Satisfaction with Nursing Care and Work scale scores), before and after the implementation—and compare with ratings from reference wards. A quasi-experimental and prospective, pretest–post-test design was used. Specific questions about the nursing programme were answered by staff at the intervention wards. Staff reported predominantly positive experiences of the nursing programme. At follow-up, higher ratings were reported in two dimensions of quality of care in the STR group, and lower ratings in one dimension of stress were evident in the reference group. No differences in ratings between the STR and reference wards were found. Staff members' positive experiences of STR and higher ratings regarding participation and secure environment after implementation suggest that STR is a well-accepted and promising nursing programme. It is important to implement and evaluate recovery-oriented interventions in psychiatric inpatient care, where a focus on symptom relief still prevails. The results indicate that there is potential for further exploration of STR in this context.
INTRODUCTION
Nursing in psychiatric inpatient care is described as peripheral in a care context dominated by the still-prevailing biomedical model (Bladon 2017). Psychiatric inpatient care is commonly viewed as aiming only to help consumers with stabilization and symptom relief and nursing is marginalized and seen as merely assisting medical decisions (Harrison et al. 2017; Holmberg et al. 2018; Lakeman & Molloy 2017; Waldemar et al. 2019). Organizational structures, such as unsupportive ward environments, can contribute to the endurance of this view of nursing (McAllister et al. 2021; Waldemar et al. 2019), as can inadequate staffing levels and lack of more than basic education in healthcare and nursing among staff members (Salberg et al. 2019; Staniszewska et al. 2019). Important and lasting changes having a significant impact on the lives of consumers can be achieved through nursing interventions, actions and approaches based on theory and research (Gabrielsson et al. 2020; Santangelo et al. 2018).
In recent years, recovery-oriented practices have become a more integrated part of nursing in psychiatric inpatient care. These practices focus on each consumers individual recovery process, not only mental illness and symptoms as in medical models of care (Chester et al. 2016). There are two overarching perspectives that can be seen in this context: clinical recovery and personal recovery (Collier 2010). From a nursing perspective, the focus is on the latter (Barker 2001; Gabrielsson et al. 2020), with personal recovery described as a personal process of change (Barker 2001; Collier 2010) and seen as a unique and individual process involving change and development. Personal recovery comprises changes of attitudes, feelings, values, goals and skills, crucial to reclaiming oneself and living a satisfying life despite limitations in everyday life (Anthony 1993). Nursing in psychiatric inpatient care fulfils an important role in an individual consumer’s recovery (Gabrielsson et al. 2020; De Ruysscher et al. 2020; Salberg et al. 2018). Recovery-oriented practices within this context imply differentiation from medical models and a shift from focusing on the diagnosis to focusing on the consumers as individuals. The care relationship and openness to consumer experiences are crucial in recovery-oriented practices (Chester et al. 2016). Clinically, such practices require collaboration, participation and practical support to prevent worsening of the consumer’s situation (Chester et al. 2016; De Ruysscher et al. 2020). Implementation of recovery-oriented practices, models of care and programmes in psychiatric inpatient care has been reported to be feasible, albeit challenging (Coffey et al. 2019; Lorien et al. 2020; McKenna et al. 2014; Waldemar et al. 2016). If there is a lack of structure and support for individualized care, nursing in this acute context will be reduced to managing risk of harm and difficult behaviours (Frauenfelder et al. 2013; Mullen 2009). Working under such conditions with low professional autonomy can provoke feelings of insufficiency in relation to consumers’ nursing needs and can lead to stress (Hylén et al. 2018; O’Connor et al. 2018). The organization has been identified as an important aspect of a healthy work environment for nursing staff and thus as crucial to quality of care (Tuvesson & Eklund, 2017). When the organizational structure and practice environment support adequate nursing measures, proactive approaches have been identified among nurses and staff (McKenna et al. 2014; De Ruysscher et al. 2020; Simpson et al. 2017).
Steps towards recovery—a recovery-oriented nursing programme
One example of a recovery-oriented nursing programme is Steps Towards Recovery (STR), focusing on the individual recovery process while taking advantage of the benefits of a group format. The programme was first developed in a rural hospital setting in Sweden (Folke et al. 2015b) and is based on behavioural activation (Folke et al. 2015b; Salberg et al. 2018). Behavioural activation is a psychological intervention aiming to reduce rumination and avoidance behaviour and increase consumer participation in activities that promote positive reinforcement (Kanter et al. 2010). In STR the focus is on developing new strategies that can be used to regain control over life and health. Participation, a central aspect of recovery (Barker, 2001) are the outset for STR and based on each consumer’s resources and needs. Empirical studies provide evidence indicating that nursing interventions based on behavioural activation are beneficial for consumers in psychiatric care, such as increased quality of life, life style changes, higher activity levels and decreased depressive symptoms (Banducci et al. 2013; Clignet et al. 2012a, 2012b; Folke et al. 2015a, 2015b; Gollan et al. 2014; Snarski et al. 2011). Results from a previous study indicate that both consumers and nursing staff consider STR as feasible and leading to improved quality of care (Salberg et al. 2018). The STR material comprises work sheets, a diary for consumers and a manual for nursing staff. By adhering to the manual in every group session, nursing staff, in their roles as group leaders and auxiliaries, facilitate the consumer’s own decisions regarding actions in their individual recovery processes.
Implementation of STR at a psychiatric clinic in a university hospital
Before implementation, the STR material was further developed together with a user representative employed at the hospital clinic regarding feasibility, wording and comprehensiveness from a consumer perspective. See Figure 1 for the structure of the programme. The implementation of STR was then designed in several steps. First, one or two staff members at each intervention ward were given the role as STR coach, underwent the education and then guided and supported fellow staff members in their work with STR, as group leaders or auxiliaries. Second, all staff members at the five wards took part in the education, divided into four two-hour blocks, led by two experienced nurses specialized in psychiatric care. The education covered theoretical and practical activities: the theory of recovery as a personal process, the themes and material as well as training in leading STR activities. Third, all coaches shared reflections, tips and strategies at joint meetings, twice a month and ward managers met once a month. The project leader and a user representative attended all these meetings. Lastly, to promote commitment to the programme, a web-based education was designed to be used as a resource and introduction for new employees.
Aims
The aim of this study was to explore the experiences of the recovery-oriented nursing programme, STR, among nursing staff, as well as their ratings of stress, quality of care and satisfaction with nursing care and work, before and after implementation. A second aim was to compare these results with the ratings of nursing staff on wards where the nursing programme was not implemented.
METHOD
Design
A quasi-experimental and prospective pretest–post-test design was used. The Revised Standards for Quality Improvement Reporting Excellence (SQUIRE 2.0) was used as a guideline for reporting the study (Ogrinc et al. 2016).
Setting
The study was conducted at eight psychiatric inpatient wards of a university hospital in Sweden. Wards were sub-specialized in psychotic disorders, mood disorders, pain-related conditions, substance use disorders and forensic care. The wards accepted both voluntary and compulsory admissions. Both males and females over the age of 18 years were being cared for at all of the wards, where the total number of beds was 73. The programme was implemented at five of the eight wards. The three wards not implementing the nursing programme were used as reference wards in this study, to enable comparisons of outcomes between STR and reference wards.
Participants
All nursing staff at the wards—nurses specialized in psychiatric care, registered nurses and nursing assistants—were invited to participate. Eligible participants included 130 nursing staff at baseline and 128 at follow-up. Due to organizational changes, one of the reference wards had been closed at the time of follow-up.
Outcome measures
Experiences of working with the nursing programme were assessed through a total of nine questions designed for the study. Questions assessed staff members’ earlier experiences of being a group leader, their use of STR and overall experiences of the nursing programme. Staff members were also asked to rate their experiences of the contents of the programme, the material, their experiences of integrating the intervention into nursing care and its contribution to a positive development of nursing care at the wards. Three of the questions where multiple choice and six were answered on a six-point Likert scale (1–6), ranging from ‘very bad/negative’ to ‘very good/positive’. This questionnaire was filled out only by staff members at the intervention wards.
Stress was assessed using the Maslach Burnout Inventory (MBI). It is a 22-item instrument divided into three sub-scales, each assessing one dimension of burnout: Emotional exhaustion, Depersonalization and Personal accomplishment (Maslach & Jackson, 1981). The questions are answered on a seven-point Likert scale, ranging from ‘never’ (0) to ‘daily’ (6). Higher ratings on Emotional exhaustion and Depersonalization indicate higher levels of burnout, whereas lower ratings on Personal accomplishment indicate higher burnout. Reliability and validity for the 22-item MBI version have been established in several countries, in mental health/psychiatric settings as well as in other professional settings (Maslach & Jackson, 1996; Paris & Hoge, 2010; Poghosyan et al. 2009).
Quality of care was assessed with the Quality of Psychiatric Care - Inpatient staff (QPC-IPS) scale. It is part of a family of instruments originating from QPC (Schröder et al. 2010). It comprises 30 items divided into six dimensions: Encounter, Participation, Support, Secluded environment, Secure environment and Discharge. Items are answered on a four-point Likert scale, where answers range from ‘disagree’ (1) to ‘totally agree’ (4). Psychometric tests are ongoing; however, some results have already been published, indicating that it is a valid and useful instrument (Lundqvist et al. 2019).
Satisfaction with nursing care and work was assessed using the Satisfaction with Nursing Care and Work scale (SNCW), which comprises 34 statements. Responses are given on a five-point Likert scale, ranging from ‘fully disagree’ (1) to ‘fully agree’ (5). The responses result in a total score between 34 and 170 (Hallberg et al. 1994). Satisfactory reliability and validity estimates have been reported for SNCW (Brodaty et al. 2003).
Data collection
Baseline data were collected at all eight wards from 18 April to 31 May 2017. The implementation of STR started in five wards in October 2017. Follow-up data were collected at both STR and reference wards from 3 February to 17 April 2020. At both baseline and follow-up, booklets with questionnaires and consent forms were distributed to all nursing staff members at the wards. Signed written consent forms were collected by the researchers. The filled-out anonymous questionnaires were placed in unmarked envelopes and left in a box, which was emptied by the researchers.
Ethical considerations
Ethical approval was obtained from the Ethical Review Board (2016/414 and 2019-04315). Being asked to answer questions about one’s work can be perceived as putting oneself at risk of being reprimanded for statements that can be seen as critique towards the organization or employer. As employees are dependent on their employing organization, they might feel obligated to remain loyal and, therefore, be reluctant to answer questions regarding their work. Ensuring anonymity is thus of great importance and the reason these questionnaires did not include code numbers.
Analyses
Due to small groups and skewed distributions, non-parametric analyses were used. Differences within the groups, as well as between the groups (STR and reference wards) at baseline and follow-up were assessed with the Chi-square test for categorical variables and the Mann–Whitney U-test for continuous variables. Statistical significance was accepted as P < 0.05. Since data were non-parametric, Cohen’s d could not be used to calculate effect sizes; instead, effect sizes (r) for changes in the respective groups were calculated using Rosenthal’s r (Rosenthal & Rubin, 2003). Limits for effect sizes for non-parametric tests were set as follows: small (0.1), medium (0.3) and large (0.5) effect (Cohen, 1988). All analyses were performed using the statistical package IBM-SPSS 27.0 (IBM-SPSS Inc., Chicago, IL, USA).
RESULTS
The results are based on a total of 145 assessments (75 pre and 70 post), with 89 from the STR wards (49 pre and 40 post) and 56 from the reference wards (26 pre and 30 post). The assessment scales were completed by over half of the nursing staff at the wards (58% pre and 55% post). There were no differences between the two groups, except at baseline, where staff at the reference wards reported longer experience in their profession and in psychiatric care than staff at the intervention wards. Due to staff turnover and organizational changes, the staff members at baseline and follow-up were not exactly the same individuals. The participants at reference wards reported longer experience in their profession and in psychiatric care at baseline compared with a follow-up. Demographic characteristics of the participants are presented in Table 1.
| Baseline | Follow-up | Baseline–Follow-up | Baseline–Follow-up | |||||
|---|---|---|---|---|---|---|---|---|
|
STR-ward n = 49 |
Reference ward n = 26 |
P-value* |
STR-ward n = 40 |
Reference ward n = 30 |
P-value* | STR-ward | Reference ward | |
| n (%) | n (%) | n (%) | n (%) | P-value* | P-value* | |||
| Sex | 0.493 | 0.455 | 0.785 | 0.445 | ||||
| Male | 15 (30.6) | 10 (38.5) | 13 (32.5) | 8 (27.7) | ||||
| Female | 34 (69.4) | 16 (61.5) | 26 (65.0) | 21 (70.0) | ||||
| Other | – | – | – | 1 (3.3) | ||||
| Employment | 0.177 | 0.157 | 0.989 | 0.173 | ||||
| Permanent | 43 (87.8) | 26 (100) | 35 (87.5) | 27 (90.0) | ||||
| Substitute | 5 (10.2) | – | 4 (10.0) | – | ||||
| Hourly | 1 (2.0) | – | 1 (2.5) | 2 (6.7) | ||||
| Professional role | 0.157 | 0.172 | 0.416 | 0.079 | ||||
| Specialist nurse | 4 (8.2) | 3 (11.5) | 4 (10.0) | 7 (23.3) | ||||
| Registered nurse | 11 (22.4) | 2 (7.7) | 7 (17.5) | 8 (26.7) | ||||
| Nursing assistant | 33 (67.3) | 21 (80.7) | 29 (72.5) | 15 (50.0) | ||||
|
M (SD) (range) |
M (SD) (range) |
M (SD) (range) |
M (SD) (range) |
|||||
|---|---|---|---|---|---|---|---|---|
| Age |
44.7 (14.9) (21-69) |
48.7 (12.4) (29-65) |
0.255 | 47.4 (13.3) (26-69) |
42.9 (16.1) (19-65) |
0.258 | 0.415 | 0.215 |
| Years in professional role |
14.7 (14.9) (0-43) |
20.6 (13.0) (1-43) |
0.043 |
17.3 (13.3) (1-46) |
14.1 (15.8) (0-49) |
0.098 | 0.184 | 0.045 |
| Years in psychiatric care |
13.3 (14.5) (1-47) |
17.4 (10.2) (2-36) |
0.024 |
11.7 (12.6) (0-44) |
12.0 (13.9) (0-49) |
0.634 | 0.924 | 0.018 |
- * Chi-square test for categorical variables and Mann–Whitney U-test for continues variables.
Experiences of the nursing programme at the STR wards
Over half of the nursing staff members at the STR wards had been involved in the programme (n = 24/40), either as auxiliaries (n = 9), group leaders (n = 5) or both (n = 10). Their overall experience of the intervention was rated 3 or higher (scale 1–6), and most of them rated it 5 of out 6 (Md 5.0, range 0–6). Not tabulated.
All nursing staff at the STR wards (n = 40), even those not directly involved in the programme, rated their experience of the nursing programme (scale 1–6). The contents of the programme were regarded as good (Md 4.0, range 2–6), as were the manual (Md 4.0, range 2–6) and the consumer material (Md 4.0, range 2–6). Staff members rated their experience of integrating the intervention into nursing care as positive (Md 5.0, range 1–6) and indicated that STR, to a large extent, contributed to a positive development of nursing care at the wards (Md 5.0, range 1–6). Not tabulated.
Stress, quality of care and satisfaction with nursing care and work
Ratings at baseline and follow-up
There were no statistically significant differences in staff members’ ratings of stress (MBI), quality of care (QPC-IPS) or satisfaction with nursing care and work (SNCW) between STR and reference wards either at baseline or follow-up. However, staff members’ rating scores for many of the domains of quality of care (QPC-IPS) where higher at the reference wards compared to STR wards before and after implementation, although the differences were not statistically significant. See Table 2.
|
Outcome measure Dimensions |
STR wards Md (range) |
Reference wards Md (range) |
U-value‡ | P-value |
|---|---|---|---|---|
| Stress—MBI | ||||
| Emotional exhaustion * | ||||
| Pre | 13 (2–47) | 21 (5–40) | 383.5 | 0.082 |
| Post | 14 (0–35) | 12 (0–33) | 502.5 | 0.247 |
| U-value | 811.0 | 142.5 | ||
| P-value | 0.433 | <0.001 | ||
| r | 0.09 | 0.50 | ||
| Depersonalization * | ||||
| Pre | 4 (0–20) | 4 (0–20) | 494.5 | 0.764 |
| Post | 3 (0–13) | 2 (0–11) | 468.0 | 0.114 |
| U-value | 803.0 | 247.0 | ||
| P-value | 0.390 | 0.076 | ||
| r | 0.09 | 0.24 | ||
| Personal accomplishment † | ||||
| Pre | 36 (17–48) | 37 (26–46) | 441.5 | 0.323 |
| Post | 36.5 (23–48) | 37.5 (20–48) | 540.0 | 0.476 |
| U-value | 800.0 | 317.5 | ||
| P-value | 0.378 | 0.621 | ||
| r | 0.10 | 0.07 | ||
| Quality of care—QPC-IPS | ||||
| Encounter | ||||
| Pre | 24 (12–32) | 25 (14–32) | 537.0 | 0.386 |
| Post | 26 (16–32) | 27 (17–32) | 472.0 | 0.170 |
| U-value | 907.5 | 322.0 | ||
| P-value | 0.686 | 0.368 | ||
| r | 0.04 | 0.12 | ||
| Participation | ||||
| Pre | 21 (13–31) | 23 (14–31) | 518.0 | 0.279 |
| Post | 24 (8–33) | 22 (8–28) | 480.0 | 0.203 |
| U-value | 722.5 | 333.0 | ||
| P-value | 0.050 | 0.477 | ||
| r | 0.21 | 0.10 | ||
| Support | ||||
| Pre | 14 (5–16) | 13 (8–16) | 602.5 | 0.908 |
| Post | 14 (6–16) | 15 (8–16) | 535.5 | 0.540 |
| U-value | 837.5 | 307.5 | ||
| P-value | 0.314 | 0.244 | ||
| r | 0.11 | 0.16 | ||
| Secluded environment | ||||
| Pre | 10 (5–12) | 10 (3–12) | 571.5 | 0.634 |
| Post | 10 (7–12) | 9.5 (6–12) | 505.5 | 0.329 |
| U-value | 930.5 | 369.5 | ||
| P-value | 0.831 | 0.925 | ||
| r | 0.02 | 0.01 | ||
| Secure environment | ||||
| Pre | 7 (4–11) | 8 (6–16) | 459.0 | 0.071 |
| Post | 8 (5–12) | 9 (4–11) | 520.5 | 0.425 |
| U-value | 528.5 | 302.0 | ||
| P-value3 | <0.001 | 0.0211 | ||
| r | 0.39 | 0.17 | ||
| Discharge | ||||
| Pre | 10 (6–16) | 12 (6–16) | 448.0 | 0.059 |
| Post | 12 (3–16) | 12 (4–16) | 459.0 | 0.167 |
| U-value | 772.0 | 332.5 | ||
| P-value | 0.170 | 0.469 | ||
| r | 0.15 | 0.10 | ||
| Satisfaction with nursing care and work—SNCW | ||||
| Pre | 70 (30–94) | 71 (41–109) | 557.0 | 0.718 |
| Post | 69 (34–115) | 66 (54–116) | 582.0 | 0.971 |
| U-value | 780.5 | 276.0 | ||
| P-value | 0.238 | 0.094 | ||
| r | 0.13 | 0.23 | ||
- Abbreviations: MBI = Maslach Burnout Inventory; QPC-IPS = The Quality of Psychiatric Care—Inpatient staff; r = Effect size, Rosenthals’ r; SNCW = Satisfaction with Nursing Care and Work scale.
- Bold values indicate p≤ 0.05.
- † High rating = higher risk of burnout.
- ‡ Low rating = higher risk of burnout.
- § Mann-Whitney U-test.
Changes in ratings from baseline to follow-up
At the STR wards, no changes were evident regarding stress (MBI). In two of the dimensions of quality of care (QPC-IPS), Participation and Secure environment, higher ratings were found at follow-up, with small (0.21) and medium (0.39) effect sizes respectively. There were no changes regarding satisfaction with nursing care and work (SNCW). See Table 2 (effect sizes not tabulated).
On the reference wards, a decrease in one dimension of stress (MBI), Emotional exhaustion, was evident, with a large effect size of 0.50. A decrease in the dimension Depersonalization, not statistically significant, with a small effect size of 0.24, was found. No changes were evident in this group regarding quality of care (QPC-IPS). In ratings of satisfaction with nursing care and work (SNCW), a small decrease, not statistically significant, with a small effect size of 0.23, was found. See Table 2 (effect sizes not tabulated).
DISCUSSION
This study aimed to explore the experiences of a recovery-oriented nursing programme among nursing staff, as well as their ratings of stress, quality of care and satisfaction with nursing care and work, before and after implementation. As a reference for clinic-wide changes over time, wards not implementing the nursing programme were included in the study.
The results indicated that over half of the staff members at the STR wards were actively involved in working with the nursing programme. They predominantly reported positive experiences, as in previous research on STR (Salberg et al. 2018), where the programme was evaluated in a similar clinical context but in a smaller local hospital setting. Nursing staff reported positive experiences of the contents and integration of the programme into nursing practice and the contribution to improvement of nursing care was rated as high. These results are in line with previous studies, where positive experiences of implementation of recovery-oriented practices in psychiatric inpatient care have been reported (Lorien et al. 2020; Salberg et al. 2018; Waldemar et al. 2016).
Nursing staff at STR wards reported an increase in their ratings of quality of care in the dimensions Participation and Secure environment. Because of the study design and sample, no conclusions can be drawn regarding any direct cause-effect relationship. However, the reports on increased consumer participation are a positive outcome, based on the direct link to theoretical perspectives for recovery-practice (Chester et al. 2016) and for nursing in psychiatric care (Barker, 2001; Peplau, 1991). Hornik-Lurie et al. (2018) reported that a positive change in staff members’ perception of consumer participation was evident after training in recovery-oriented practices. Furthermore, staff members’ experiences of consumer participation correspond to previously presented results of consumers reporting an increased involvement in care and planning after participation in a nursing intervention similar to STR (Salberg et al. 2018). A programme like STR is an important factor in targeting social disengagement and passivity among consumers, previously described in this clinical context (Folke et al. 2018; Lindgren et al. 2015; Molin et al. 2016; Sharac et al. 2010). Care at a psychiatric ward cannot be a demarcated zone in health care, change is needed. Person-centred support and possibilities to begin and continue the individual recovery process should be provided. The strong biomedical perspective in psychiatric care may aggravate initiatives to adapt to a more person-centred care, since reduction of symptoms often are considered the primary outcome. Hence, a psychiatric inpatient care where recovery-oriented nursing, based on evidence, is self-evident and integrated into nursing practice is crucial. Recovery-oriented practices like STR can give emphasize to the importance of nursing and thereby be a foundation for adaptation to person-centred care.
The reports from nursing staff of a positive change in Secure environment could be regarded as an important aspect of enabling the persons being cared for to feel safe. Hence, STR may contribute to creating a more predictable and meaningful nursing care. In previous research, a predictable treatment and care process has been described as necessary for consumers to feel safe in psychiatric inpatient care (Pelto-Piri et al. 2019). Safety is described as an important prerequisite for recovery in psychiatric care by both consumers (De Ruysscher et al. 2020; Staniszewska et al. 2019) and staff (De Ruysscher et al. 2020). From a consumer perspective, experiences of safety can be enhanced by nurses supporting of personhood and treatment as an equal, instead of nursing solely focusing on risks (Cutler et al. 2020). Furthermore, focusing on recovery in nursing care can be utilized in strategies to reduce the risk of aggressive behaviour (Lim et al. 2017). In conclusion, nursing focusing on recovery can support maintenance of a safe environment in the psychiatric wards (Lim et al. 2019), also indicated by the results of this study.
There were no changes from baseline to follow-up in stress or satisfaction with nursing care and work at the STR wards. This is in contrast to results in a study where staff at psychiatric inpatient wards, after implementation of a recovery-oriented concept, rated satisfaction with working conditions and contentedness, as well as ward atmosphere, higher than a control group (Rabenschlag et al. 2014). At the reference wards, there was a decrease in one of three dimensions of stress, Emotional exhaustion. A tendency towards a decrease in the dimension of Depersonalization was also indicated, though it was not statistically significant. It would be interesting to consider what might have led to these changes over time, but as this was a naturalistic clinical study, we can only speculate on that. Furthermore, the staff members at baseline and follow-up were not exactly the same individuals. One difference was that staff members at the reference wards had longer experience in their profession and in psychiatric care at baseline than at follow-up. In previous research, higher stress levels have been linked to senior staff having a heavy responsibility for consumer well-being (Johansson et al. 2013). However, other results have indicated less stress (Foster et al. 2021) in more experienced nurses thanks to the development of resilience skills (Foster et al. 2020).
As discussed, there were no statistically significant differences between STR and reference wards in any of the outcome measures. However, it is interesting to consider possible reasons why ratings from nursing staff at the reference wards were higher before and/or after implementation in many of the domains of quality of care (though the differences were not statistically significant). Wards were enrolled in the implementation of STR based on the decision of the ward manager; it is possible that the nursing care at wards not enrolled was already of high quality, explaining the higher ratings. Another possible explanation might be that knowledge about the implementation at the STR wards led to a deeper, clinic-wide understanding of the need to focus more on the individuals’ recovery in nursing practice, leading to lower ratings of quality of care.
The identified changes within the groups are not easily explained solely by the implementation of the nursing programme, as no conclusions can be drawn on cause-effect relationships. The lack of decrease in stress in the STR group may be tentatively explained by the initial stress of working with new methods, as discussed by Koivunen et al. (2013). It could also be argued that the implementation of STR was too small a change to have an effect on nursing staff members’ satisfaction with nursing, or that other aspects of the working environment affected the ratings. As described earlier, forming and maintaining care relationships is a crucial component of recovery-oriented practices. However, transitioning from an advisory or illness expert role to a collaborative relationship is difficult and the culture and conventions of a prevailing biomedical model can be intractable (Chester et al. 2016). Some previously identified challenges when implementing recovery-oriented practices include organizational logistics, competing demands and care culture (Cusack et al. 2017; Hornik-Lurie et al. 2018; Lorien et al. 2020; McAllister & McCrae 2017; Waldemar et al. 2019). In a recent review, the authors concluded that there were several factors impeding implementation of mental health recovery into services and for transformation to happen there was a need for more general changes across the whole organization (Piat et al. 2021).
Limitations
More distinct results could have been achieved with intragroup analyses within the subgroups. However, the number of participants was low and due to staff turnover and organizational changes, the participants before and after implementation were not exactly the same. Hence, such analyses were not feasible. The wards enrolled in this study were all sub-specialized, but differences between the wards were not explored further in this study. Further, the coronavirus pandemic led to strained working situations at the wards during implementation, which might have affected the ability to work in accordance with STR.
The consumer perspective was not represented in this study, hence there is an obvious lack of consumer-focused evidence for STR. There are, however, ongoing evaluations where the experiences of consumers who has taken part of STR is in focus, both regarding the programme and how it has affected their care and recovery process.
CONCLUSION
Nursing staff predominately reported positive experiences of STR. Moreover, consumer participation and secure environment, both important aspects of the individual recovery process, were rated higher after implementation. These results are promising and indicate that there is potential for further exploration of STR as a recovery-oriented programme in psychiatric inpatient care, focusing on the individual recovery process while taking advantage of the benefits of a group format. The programme itself can, however, be seen as a contribution to enhance the autonomy of nursing in psychiatric care and creates a forum for empowering consumer’s personal recovery. The results show that is it possible to work with a focus on the individual recovery process without waiting for medical care decisions from the physician. Stating the importance of autonomous nursing in the context of psychiatric inpatient care is needed to achieve high quality care.
Relevance for Clinical Practice
Recovery, as a personal process of change, is a core concept for nursing in psychiatric care. Implementation of recovery-oriented nursing into psychiatric inpatient care enables incorporation of theory and evidence-based practice into clinical nursing. It can contribute to higher quality of care through enabling nursing with focus on the resources and needs of the individual consumer. Hence, support and guidance for nursing staff are crucial, as refocusing from an advisory or expert role to a more collaborative nursing care can be demanding, especially in a culture where nursing is described as peripheral and the biomedical model of care is dominant.
Acknowledgement
We would like to thank Fredrik Folke for designing the initial programme and for inspiration and valuable contribution when further developing the programme and planning for implementation.
Ethical Approval
Ethical approval for the study was obtained from the Ethical Review board (dnr: 2016/414 and 2019-04315).
FUNDING STATEMENT
This study received funding from Märta and Nicke Nasvell’s Grants in psychiatric research.
