Advancing the Recovery Orientation of Hospital Care Through Staff Engagement With Former Clients of Inpatient Units

Developing recovery-oriented services is the goal of modern mental health care (1). In the context of severe mental illness, recovery is a process that helps people “liv[e] a satisfying, hopeful, and contributing life even with limitations caused by the illness” (2). Recovery-oriented care, in turn, supports clients as they pursue their own goals by providing care that is individually tailored, respectful of rights, and strengths based and that promotes consumer involvement and hope (3).

Efforts to develop and invest in recovery-based approaches, such as supported employment (4) and Illness Management and Recovery (5), have focused on outpatient services. However, the inpatient setting often represents a critical juncture for people with severe mental illness. For many, it is their first substantive point of contact with the mental health care system, and most access inpatient care on multiple occasions during the course of their lives (6). In Canada, for example, approximately 75% of people with schizophrenia have been hospitalized, 38% are readmitted within one year, and the average length of stay is 27 days (7,8).

Several studies have examined the challenges involved in implementing recovery-oriented care on inpatient units. Examination of client perspectives have highlighted that effective engagement by inpatient staff can have a major impact. Conversely, experiences of coercion, threats to safety from staff and other patients, and a general process of dehumanization can completely compromise care at a time of crisis (9,10).

Broader commentary on the topic has suggested that wards have difficulty applying recovery principles to their work for at least three reasons. First, delivering care driven by the agenda of a vulnerable, distressed patient who may be detained against his or her will may seem to run counter to ward practices that increasingly emphasize risk management and clinician-driven decision making in short-stay frameworks (9,11). Second, although recovery may be understood at a conceptual level, it is seldom applied because clinicians are not educated to develop the skills and perspectives necessary to support such an orientation (12–14). Often lacking a coherent model, the treatment context compounds this problem by a high degree of variability across staff and unit practices (15). And last, the very nature of the staffing experience on inpatient units represents a challenge, given that staff often have exposure to clients only when they are in crisis and not when they are engaged in work, in school, and in the community. This limited exposure has an impact on their expectations for clients and can undermine their belief in recovery-based care (12).

Narrative-based teaching is a possible tool for broadening the experience of inpatient staff. Narratives have long been considered an important tool for exploring, learning about, and developing shared understandings between service users and organizations across many care sectors, including amputation, stroke rehabilitation, and arthritis (16–20).

Extensive general commentary agrees on the importance of narratives by consumers to mental health reform. These narratives have played a key role in decades of consumer advocacy (21–23), and they appear to have an impact on attitudes among general audiences (24). However, studies of the impact of narrative-based teaching on service providers are scarce (21,25). There has been no study of the effectiveness of narratives on improving staff attitudes, skills, or knowledge related to recovery-oriented care.

We developed a study to assess the impact of consumer narratives on inpatient service providers. We invited individuals who had previously been inpatient clients on schizophrenia inpatient units to deliver a yearlong series of talks to staff of the units. It was hypothesized that exposure to the talks would affect the providers’ knowledge and attitudes about recovery-oriented care. We also probed the impact of this strategy on the implementation of recovery-oriented care at a unit level. Last, we examined whether a more optimistic understanding of the possible outcomes for inpatients might improve job satisfaction.

This study was intended to address a gap in policy, practice, and research literature. The importance of drawing upon consumer narratives and input to improve services is increasingly emphasized, but there is very limited study of their impact. The study also comes at a time when consumer advocacy groups are critically examining how consumer narratives are used and raising questions about the potential for narratives to be co-opted and placed in frameworks that serve to reify custodial practices (26).

Methods

Setting, design, and analysis

This study took place in the schizophrenia program of a large psychiatric treatment facility located in a Canadian urban center. This service provides extensive outpatient services and includes six inpatient units with an even dispersion of approximately 120 beds. Staffing is interprofessional, although made up primarily of nurses, and each unit is staffed by approximately 15 regular staff members.

The intervention was a series of talks given by former patients twice a month over a one-year period between May 2011 and May 2012. To examine the effectiveness of the speaker series, a mixed-methods approach was taken. The six inpatient units were paired on the basis of the characteristic length of stay (short, medium, and long), with one ward randomly chosen from each pair to get the intervention.

Self-report measures completed before and after the talks assessed knowledge and attitudes regarding the recovery model at an individual level, delivery of recovery-oriented care at a unit level, and job satisfaction. In addition, unit staff participated in focus groups about the talks, and individual interviews were held with the speakers themselves. The study was reviewed and approved by an institutional research ethics board, and informed consent was obtained from all participants.

Quantitative assessments of the impacts of the talks were examined with analyses of variance (ANOVAs). Qualitative interviews were audio recorded and transcribed verbatim. An investigator analyzed the transcripts by using a content analysis strategy (27), in which recurrent categories were identified through line-by-line coding. The code structure was then examined by a second investigator, and the two investigators negotiated some minor revisions. Together they identified cross-cutting themes.

Measures

The Recovery Self-Assessment (RSA) was used to measure the recovery orientation of care, with staff instructed to consider items as they applied to their specific unit (3). Five subscales are embedded in the 36-item tool: facilitation of pursuance of individually defined life goals, involvement of consumers in service provision, diversity of treatment options, respect for rights and choice, and provision of individually tailored services. The tool has demonstrated good validity and reliability in previous work (28).

Personal attitudes and knowledge regarding recovery-oriented practices were assessed with the 20-item Recovery Knowledge Inventory (RKI), which includes four subscales: roles and responsibilities in recovery, nonlinearity of the recovery process, the roles of self-definition and peers in recovery, and expectations regarding recovery. This scale likewise has demonstrated good psychometric properties (29).

To evaluate job satisfaction, the 16-item Job Satisfaction Scale (JSS), a well-validated tool for use in health service provision, was used (30).

By necessity, no demographic information was gathered from staff. Staff members expressed a great deal of sensitivity about reporting their practices and perspectives on the above domains and, therefore, requested complete anonymity.

Intervention

The target was to provide talks on each of the three intervention units twice per month (to facilitate access for different shifts), for a total of 72 talks. Because of unexpected staff shortages, speaker illness, or unit emergencies, a small number of talks were canceled. A total of 58 (81%) of the scheduled talks were completed. The average attendance was four or five inpatient staff per talk and eight or ten staff per month, out of 15 regular staff whose schedule made it possible to attend. Talks were given in a meeting room on the unit, were 20 minutes long, and were followed by five to 10 minutes of questions and conversation. Attendance at the talks was mandatory, and managers adjusted schedules to allow staff to attend and encouraged their attendance. However, unit managers very seldom were required to employ interventions to enhance attendance, given that staff demonstrated a strong interest in the project.

The speakers were 12 individuals with a history of psychosis who had previously been treated as inpatients at the institution. The criteria for selection were having successfully engaged in their recovery, having demonstrated some skill in articulating their experience, and having been comfortable with giving talks in potentially stressful environments. The resulting group of 16 prospective speakers was diverse in age (mid-twenties to mid-sixties) and race-ethnicity, had equal numbers of men and women, and represented a range of recovery experiences. They included individuals who were part-time volunteers, university graduates, people who were employed, and people who were not employed. Their perspectives ranged from feeling deeply appreciative to highly critical of the care that they received.

Speakers attended two half-day workshops in which they were oriented to the project, with group exercises undertaken to begin to formulate their talks. Speakers then met one on one with hospital peer support workers, who acted as mentors, between two and five times to further develop and practice their talks. Of the 16 candidates, three did not complete the training because of difficulties in creating a talk or attending training meetings, and one gave one talk but then withdrew after experiencing stress related to being back on the ward. Their talks included a diverse range of recovery-related experiences—for example, working, education, volunteering, and art—and the resources they draw upon to facilitate recovery—such as support from family and staff, creativity, spirituality, and appropriate medication. They also gave feedback on experiences they had in the units, including positive experiences with staff, such as staff taking time to get to know them personally and expressing care and empathy, and negative experiences, such as insulting behavior by staff, initiation of seclusion for inadequate reasons, and bullying by other patients.

Results

Quantitative findings

Surveys were completed by 37 staff on the intervention units and 22 staff on the control units before the intervention and by 22 staff on the intervention units and 38 staff on the control units after the intervention. Approximately 90 regular staff work on the units. Survey respondents on the intervention units indicated having attended an average of three talks (ranging from two to four). Our observations of staff attendance would suggest that these respondents represented a midrange of all staff, given that a small number of unit staff attended nearly every talk, and a small number attended only one talk.

Findings for the RSA, RKI, and JSS before and after the intervention are reported in Table 1. The ANOVA findings revealed no significant effects for RSA and JSS total scores. Similarly, there were no significant findings for RSA subscale scores. However, there was a significant interaction effect for the RKI total score (p=.03). Broken down by subscale, this interaction effect seemed to be explained mainly by scores on the subscale measuring nonlinearity of the recovery process, the only subscale with a significant interaction effect (F=6.46, df=1 and 119, p=.01). The effect size for the RKI mean score fell in the moderate range (Cohen’s d=.68).

Qualitative findings

Staff focus groups.

Twenty staff across all three units attended focus groups about experiences with the speaker series. A thematic content analysis revealed two major themes. The first theme focused on the recovery narratives and the associated sense that “there is hope.” The second theme involved implications for staff in their clinical roles.

With respect to the theme of hope, staff contrasted the recovery narratives with the experience of working on inpatient units where the clients were “in crisis” or otherwise at a “low, low point,” where it was difficult to imagine “how they might thrive.” Staff described a very positive reaction to the narratives, characterized by an improved sense of “hope for clients” and “hope in the impact that we [staff] may have.” This improved feeling of hope contributed to a sense of pride in doing a job that can feel like “working all day and you don’t see anything come of it.”

Another theme that was framed in the context of hopefulness was an improved sense of empathy. Hearing about people’s lives outside the confines of the relationships on inpatient units made some staff reflect on how, in many respects, they could relate to the life experiences of clients. Finally, there was some discussion about how a better understanding of the staff’s role in the larger recovery process provided “a boost of hope and energy to come up with new and innovative ideas on how to maybe change [our] practice and thinking.” Only one staff member presented a dissenting view in this area, noting that recovery happens “somewhere out in public” and “not on the inpatient unit.”

The second major theme revolved around how the speaker series “spurred reflection” on practice. Staff described this as a rare occurrence because they were so “caught up in our jobs,” and they valued the chance to consider the implications of the narratives “before we go back into our worlds.” Central to this consideration, and consistent with messages around hope, was the understanding that clients “are human beings, too, and they are going through a tough time.”

The clinicians were asked how they validate and respond to the humanity and adversity of patients. Many staff members were struck by how the speakers regarded seemingly “insignificant things” as critical elements of recovery-facilitating care. These included having a “human” connection with staff, having staff take even a small amount of time to listen to a request or discuss a challenge, and having conversations with staff that extended beyond questions about medications and symptoms or “nursing staff telling them what to do.” Several staff described making more efforts to be responsive and seeing positive outcomes as a result. These outcomes included fewer instances of clients becoming frustrated and prompting cycles of escalation, clients becoming more empowered through programming that addresses their goals, and clinicians becoming more active in connecting clients with outpatient services. “We can’t just keep pushing meds, we have to take into account what they think works for them, too.”

Staff also highlighted the many constraints, particularly those related to documentation, that challenge their ability to better engage their clients. One staff member said, “Nursing staff have become much more involved. ‘Get off the unit and get to know your clients’ seemed to be the key message from all of the speaker series. That was vital to their recovery and if they had that opportunity to talk about their hopes, their dreams, their wishes, their goals . . . Everyone’s been working together a little bit better because of that general message.”

Speakers’ interviews.

The most prominent theme in the speakers’ descriptions of the project was empowerment. They described a process of growing confidence, as they successfully engaged staff at a “human level,” in effect recalibrating a relationship that had been characterized by a marked power differential. As one speaker remarked, “It was very empowering to be able to go back to the wards, the one ward in particular, in which I felt really dehumanized. It wasn’t easy to go back and see it . . . but the experience of being able to go back and give feedback, and for the most part being respected for giving feedback . . . I felt, it was really rewarding. It was kind of nice when one of them who I thought was really hard and uncaring, she actually said, ‘Thank you so much for you input. It means a lot when people make a recovery.’”

Discussion

This study employed a mixed-methods randomized design to examine the impact of the recovery narratives and feedback of former inpatient clients upon the recovery orientation and job satisfaction of inpatient staff working on schizophrenia units. It was developed to address the paucity of literature and service development tools that address efforts to advance the recovery-informed model of care in inpatient contexts (11,13,14).

The hypothesis that the speaker series would have an impact on the attitudes and knowledge of staff with respect to recovery was supported. Subscale analysis indicated that the change in thinking about mental illness lay primarily in a change in beliefs about the linearity of the recovery process. The increase in scores for this subscale indicated a greater recognition that recovery encompasses a diversity of pathways and resources. Probes into impacts on recovery-oriented care at a unit level and job satisfaction revealed no significant change. Mirroring the quantitative findings, qualitative findings reflected a greater appreciation for the potential of inpatient clients and a greater recognition of the importance of engaging clients at a “human” level by recognizing their individuality and treating them with respect. This emphasis upon “human” interactions echoes aspects of care valued by clients in other contexts (10). Although quantitative indicators did not detect a change in practice at the unit level, some providers qualitatively noted having made active efforts to be more responsive to client requests, engage in more individualized care planning and, in general, become more motivated to provide support that reflected the full potential of their clients. Some staff also noted an increased feeling of pride in their work, as they recognized their role in a more hopeful process of support.

Overall, both speakers and providers agreed that the greatest value of these interactions was the opportunity to connect outside the confines of the sharp power differentials that typically characterize provider-patient relationships. The speaker series served as an exposure strategy that had an impact on staff attitudes and knowledge, which is the first step toward substantively changing practice. It also demonstrated that skills and implementation cannot be derived through exposure alone, and a more substantive investment is needed to fundamentally change the way that inpatient care is provided (31,32). To effectively have an impact on practice, this training would likely need to be considered as a component of a larger intervention that would more explicitly address skill development through mentorship and organizational structures that embed new practices as routine care (32).

There were several limitations to the design of this study. First, we were unable to use a paired-sample strategy because of staff concerns about anonymity. Second, we are unable to comment upon the generalizability of these findings beyond the context of inpatient care in a single hospital. Third, the lower baseline score on the RKI for the intervention group may have contributed to the significance of the findings. Last, there are a very wide range of potential determinants of the variables examined in this study, and further study is needed to assess reliability.

Conclusions

Despite its limitations, this study supports the use of a consumer-engagement approach in psychiatric inpatient units, settings that so far have been largely overlooked in recovery-oriented care dialogues (11–14). Furthermore, it provides evidence for the effectiveness of a rigorous and focused patient engagement strategy in psychiatric care. Such work is much needed, given that diffuse approaches to consumer engagement that are not informed by research run the risk of being superficially employed and of obscuring meaningful efforts to advance mental health reform (26).