Persons in need of psychiatric treatment and care have been front-runners in terms of self-management of disease and taking control over their lives despite disease, as well as organizing peer support. This started in the United States as a consumer movement in the 1970s [1] and coincided with the deinstitutionalization of psychiatric services. At the same time, longitudinal studies of patients with mental disorders changed the previously pessimistic view on severe psychiatric disorders, well in line with the personal experience of people that they were, despite a psychiatric diagnosis, able to lead meaningful and productive lives. Those studies also showed that the capacity of patients to recover fully or learn to manage their condition with strategies which were in many instances developed outside formal treatment settings [1,2]. During the last 30 years, an emergence of locally based services, therapeutic societies, and the development of social psychiatry have taken place, which has generated both new knowledge and new practices [3,4].

Perkins et al [5] contrast this patient-driven self-management activity from professional psychiatric care by referring to it as an “educational” rather than “therapeutic approach[es].” Instead of focusing on problems and dysfunctions, and limiting activities to therapies, the recovery movement supports people to identify and develop their talent and skills, explore their possibilities, and focus on achieving ambitions and goals. It has, consequently, also been defined as an “assets-based” approach, aiming at developing the “recovery capital” of patients, defined as “the array of social, psychological and cultural networks beyond professional inputs” [6]. It has subsequently been seen as complementary to specialist mental health services [7].

Recovery colleges (RCs) are examples of that educational approach. The first RC was established in Boston in the 1990s, initially named Recovery Learning Center. In Europe, the first RC was established in London in 2009, and today a total of 80 RCs are spread across Great Britain. Twenty other countries have followed suit [5]. An RC can best be characterized as an educational center based in the community with the overall aim to promote recovery among persons with mental health challenges by focusing on education and support [8,9]. These colleges organize, led by patient peers, courses in, inter alia, “Living with a mental disorder” and provide participants with skills and employment training, along with actions to promote social inclusion and reduce stigma. The courses are developed and co-designed, coproduced, and cofacilitated by both mental health professionals and persons with user experience.

Recovery is said to have its root in psychiatric rehabilitation, but its proponents tend to distance themselves from mainstream psychiatry [10]. On the other hand, initiatives such as UK RCs aim at moving beyond patient involvement to coproduction and the development of coproduced patient pathways. Importantly, Topor et al [11] emphasize that the lack of “recovery capital” among patients is associated with lack of trust in oneself, and that professionals are able to support if they gain acceptance as trustful agents by cocreating changes in those patients’ lives. They also report that professionals were concerned with the fragmentation and societal shortcomings of their care organizations [11].

Recent research shows that RCs have overall positive outcomes from their activities. Several studies on students’ experiences underscore that RCs provide a safe space and meaningful connections and give room for personal growth, empowerment, and hope [10,12-14]. Students using RCs seem to be, in general, representative of mental health service users, but some groups are underrepresented, such as men, older adults, ethnic minority groups, and those with developmental disorders [15,16]. RCs display consistent values and aims in supporting recovery but are diverse in their operations [8,17]. Furthermore, RCs differ widely in organizational structure and whether they are affiliated with a health trust or a local community. Research examining RC characteristics concerning their affiliation is sparse, but a recent study comprising 36 RCs across England found that those 2 RC types did not differ significantly on either clinical or any health service use outcomes [18].

The recovery movement within mental health is a prominent example of a patient-led activity that can improve the well-being of patients and, when carried out in cooperation with mental health services, contribute to their development into more patient- and person-centered organizations. Research shows that recovery principles in mental health have the potential to improve services, but that these principles are poorly implemented [19-21]. The challenge is to make the potential partnership come true by building trust, overcoming ideological barriers, and managing resistance to change among mental health professionals and their organizations.

These are organizational and leadership issues. However, the literature on RCs has until recently been scarce on organizational studies. The English RECOLLECT project, studying primarily RC effectiveness and cost-effectiveness, set out to identify organizational influences on fidelity and improvements in mental health outcomes and found that RCs basically fall into 2 categories: those located within National Health Service (NHS) Trusts with structured recovery planning and those outside the NHS that focused on social connection and community integration. No significant differences were found between those RC types in terms of perceived participant benefit [16]. Thaki et al [18] reported on how organizational factors facilitate or hinder the establishment and management of RCs. They identified the main organizational affiliation, roughly two-thirds of RCs being affiliated with a statutory health service (usually an NHS Trust) and one-third with a voluntary organization. Managers interviewed mentioned that the benefits of being affiliated with an NHS Trust were access to qualified professionals who can participate in coproducing RC workshops, higher professional staff awareness, and increased visibility to secondary service providers supporting service users, one example being employment agencies. Negative consequences were competition for resources within the NHS Trust, lack of support to the RC as higher priority is given to clinical services, and NHS bureaucratic rules restricting autonomy for RC operators. However, these insights were provided by an analysis of all interviews without identifying possible differences in experience between managers representing different organizational forms of RCs [18].

We address this gap in knowledge by analyzing and comparing 2 RCs in 2 different countries that differ in their formal relationship with the mental health service. One has a narrow focus and is embedded in a psychiatry organization engaging salaried staff with lived experience and the other is freestanding with voluntary workers and a wide range of activities. We chose those cases as we had since several years had access to the sites and had previously performed studies on the experience of participants. Despite the study on RCs aligned either to a mental health service or to the third sector [18], to our knowledge, no study involving an RC within psychiatry has been previously reported.

An educational program, called “the Patient School” and resembling an RC, is offered at the Region Stockholm (Sweden) Psychiatry Organisation by “user-influence coordinators” (UICs) and “staff with user experience” (SUEs) employed by the organization.

Sagatun Recovery (SR) (user-led center) in Hamar, Norway, is a freestanding facility organizing a range of recovery-supporting activities, locally, regionally, and nationally, with the overall aim to promote user involvement, empowerment, and recovery. Its “self-empowerment course (SEC)” is organized as an RC.

The purpose of this study is to make a systematic comparison of the embedded versus the freestanding RC in terms of their perceived value and the views of mental health staff in those respective environments. Despite differences in content and organization these 2 RCs have the same overarching aim: to support recovery and coping by providing information and supporting self-management skills.

Both RCs have recently been evaluated by in-depth qualitative analyses. In Stockholm, course leaders, participants, and health care staff were interviewed individually with a focus on the perceived value of the course as well as promoters and barriers to implementation [22,23]. The value of Sagatun courses as perceived by participants was analyzed in focus group interviews, especially in relation to whether participants had an increased sense of independence and reduced dependence on public services, and whether the course had shown new ways of practicing user-provider collaboration [24]. In this study, we complement these findings with interviews with professional mental health staff in order to shed further light on how organizational factors, as well as their attitudes and practices, could further promote collaboration. The aim of this comparative study is to uncover whether there are differences between the embedded and freestanding RCs (1) in the perceived value of the RCs by participants, and (2) in psychiatry staff views on the benefits and challenges observed of the RCs and potential for collaboration.

Ethics approval was granted by the Regional Ethical Review Board of Stockholm (Dnr 2019‐03849 decision date October 22, 2019, with amendments Dnr 2020‐04604 decision date September 15, 2020, and DNr 2021‐04724 decision date September 20, 2021) and by the Norwegian Centre for Research Data (NSD nr 332381 decision date May 15, 2023). All procedures followed were in accordance with the Declaration of Helsinki of 1975, as revised in 2000. Written informed consent was obtained from all informants of this study.

The study design is an explanatory comparative case study. It uses the fact that the research group had access to 2 different user-led patient RCs, thus benefiting from increased variation in terms of contextual factors. The cases are from 2 countries, one being embedded in a mental health service and the other being a freestanding organization.

The case study research method is “an empirical inquiry that investigates a contemporary phenomenon within its real-life context; when the boundaries between phenomenon and context are not clearly evident; and in which multiple sources of evidence are used” [25]. The explanatory approach is used when the objective is to disclose causal links between real-life interventions and their outcomes, and experimental designs are not possible because of the complexity of events [25]. Those causal links will be demonstrated as empirical patterns, showing links between stakeholder actions, organizational and environmental conditions, and consequences observed.

This study compares how 2 RCs have been established, organized and run, and has the ambition to uncover benefits and challenges and seek explanations for those. That resembles the application of new knowledge or technologies and the introduction of new policies or practices. Greenhalgh et al [26] reviewed the literature on innovation and change in service organizations. They concluded that real and persistent change is achieved through an intricate interaction between the external environment and its stability, the properties of the innovation (change), the implementation strategy, and the actual change processes. Attention should be paid to the motives and actions of the stakeholders involved and their interrelationships, as well as the internal environment characterized by its structure, resources, leadership, and internal communication. To manage this complexity, a framework to guide data collection, analysis, and interpretation is warranted. Whipp and Pettigrew [27] proposed in their “strategic change management model” that 4 dimensions of change are studied and analyzed in relation to one another and over time: the local environment (context), the planned intervention (the content of the change), the implementation (the process of change), and outcomes, both in relation to predefined objectives and unintended consequences. This framework captures the contributing factors highlighted by Greenhalgh et al [26] but is simpler and thus a more practical tool for guiding data collection and analysis.

Individual and group interviews had been previously performed by the authors with RC participants as reported elsewhere [22-24]. Mental health staff were purposefully recruited to represent different services used by participants on both sites and interviewed individually or in groups (interview guides will be found in Multimedia Appendix 1). Documents consisting of advertisements, user information, course plans, and course material were studied. The data collection was guided by the Whipp and Pettigrew [27] framework in order to enable case descriptions with the following content:

Interview data and documents were systematically examined by conventional content analysis [28]. Interviews were transcribed verbatim. The transcripts were subjected to conventional content analysis using an inductive approach. LA-A read the Stockholm transcripts and RJR read the Sagatun transcripts several times to reach immersion. They formulated independently meaning units with relevance to the study aims and covering all 4 dimensions of the theoretical framework. Those selected meaning units were then compared and discussed by LA-A and RJR and checked against the original transcripts. When consensus was reached the meaning units were labeled, grouped into categories (forming subthemes) and ultimately themes, separately for the 2 cases. Those were discussed by all authors, MB having detailed insights into the Stockholm case and HP into the Sagatun case. Consensus was reached on the appropriateness of the theme titles and related descriptions. As the views of the informants were used as an input to the characterization of the cases, dependent on other sources as well, member checking was not felt to be necessary.

Data from all sources, including the initial interviews with RC participants, were cross-checked and condensed into 2 thick case descriptions, structured according to the analytic framework [27]. Those are presented in full in Multimedia Appendices 2 and 3. Empirical patterns demonstrating the character and interrelations of the context, content, process, and outcome items were looked for and identified as tentative explanations to the benefits and challenges observed.

The principal findings in terms of similarities and differences as well as benefits and challenges when comparing an embedded versus a freestanding RC are the following:

The main similarities between recovery activities in Stockholm psychiatry and Sagatun are that both are led by persons with lived experience, focusing on supporting users in self-managing their lives through practical information and skills training. Sharing experience is an important part of the dynamics. Users experience that the courses are empowering and have the potential to reduce the stigma of mental illness. Mental health staff mostly appreciate the activities and inform their patients about the possibility to enroll, but they have limited knowledge about the content. The professionals feel that the activities potentially can lead to a more balanced relationship between professionals and users. However, no one mentions an impact on clinical practice or care arrangements.

There are major differences between the 2 sites. Stockholm psychiatry has almost a 20-year history of employing persons with lived experience with the task of enhancing user influence. They have introduced a number of routines to strengthen the position and rights of patients. One of the initiatives was the introduction of the RC.

Sagatun is a freestanding user-run facility established over 20 years and has a broad range of low-threshold services, including offering a physical meeting place and social activities in addition to self-empowerment support. It has a formal role as a regional resource and competence center and is, as such, well known both in mental health and local communities.

The Stockholm RC is an on-off experience; Sagatun offers continuity in terms of access to the center and participation in its various activities. Despite these differences, the perceived benefits are, by and large, the same. Users tell about benefits such as improved self-management skills. Mental health staff see the potential for a more balanced relationship between professionals and users, although there is little evidence of impact on professional roles and practices. In conclusion, these differences do not seem to be of importance in relation to perceived benefits, neither by users nor by mental health staff. Although staff appreciated the RCs and acknowledged the potential for improved collaboration with users, RCs did not seem to promote clinical practice changes.

When further scrutinizing outcomes, one has to look for nuances at best. Stockholm user participants emphasized as benefits the following: a sense of belonging, exchange of experiences, and practical knowledge helping to navigate the system (of care, social services, and support). They also felt a reduced social stigma. Sagatun clients appreciated the social forum offered by the center, acquiring knowledge of patients’ rights and long-lasting contacts with peers.

To summarize, we could not see any decisive differences between the embedded RC of Stockholm and the freestanding Sagatun facility. In both cases, the general knowledge among mental health staff about the activities was scarce. However, those who knew about the courses tended to recommend them to their patients. There was no evidence in staff interviews that the courses had influenced clinical practice. At best, RCs could have an impact on the care culture, reducing a diagnosis focus and promoting client-centeredness and coproduction, and be a supplement to formal mental health services.

However, the case comparison offers potential explanations to what could contribute to a successful RC, both offering benefits to users, improving communication and collaboration between users and mental health professionals, leading to practice changes and enhanced user-centeredness.

Having persons with user experience employed as staff in a mental health service would create a companionship that could promote a better understanding among professional staff and a more balanced relationship supporting recovery. It could also promote the internal spread of recovery-supporting activities of benefit to users, also showing how it can complement and potentially reduce the need for psychiatric care.

When organizations such as Sagatun offering recovery activities receive an institutional status in the mental health field, that will increase the general understanding of their benefits, have a potential for wider outreach, and lead to mental health services actively seeking cooperation and potentially integrating those into their service strategies.

Organizing courses that offer practical advice about the service system will increase users’ ability to navigate the system and self-manage. On the other hand, a well-thought-through psychoeducational approach with trained tutors or coaches offering skills training grounded in sensemaking and connectedness will create a firmer base for users to tackle challenges in everyday life. These are recommendations applicable to RCs in general, regardless of their organizational affiliation.

An initial assumption was that there would be differences in mental health staff views on the embedded and freestanding RC. In fact, there were few differences. Staff had, by and large, a positive image of the RCs but were not well informed about course activities. Those whose patients had participated in the courses had more knowledge and appreciated the contributions of the programs. Few saw the college as an activity that could complement their professional services, and there was no indication of impacts on their clinical practice. No one raised that RCs represent an educational effort which could complement clinical interventions. One reason might be that few of the staff had actively participated in an RC—this was the case in the embedded RC also.

In fact, most studies that have assessed the influence of RCs on mental health staff and their practice have been performed among professionals participating in RCs. Many of these studies focus on the value of coproduction during college activities. According to one of the studies, staff that had participated in RCs reported that the experience had changed their practices. They had gained increased understanding of service users’ relations, realities, and experiences, and modified their assumptions and biases concerning users. In addition, they had moved to “recovery-oriented practice acknowledging user self-determination, the importance of peer support, and promoting strength-based approaches” [29]. The same was reported in 2 studies performed among NHS mental health staff [30,31]. A recent systematic review found that coproduction in RCs affected professionals’ attitudes and the power dynamics between patients and practitioners and promoted a person-centered practice [32].

The value of working together was raised by Dalgarno and Oates [33]. As a result, senior clinicians had changed their perceptions of professional power and authority. A challenge, though, was the need to balance multiple roles and to see service users as colleagues. Reflecting on their clinical practice, these clinicians said that they had adapted their language to users and shared personal information with them. Another group of psychiatrists were aware of and had positive views of the RC concept. They had, however, concerns about its interaction with the medical model. There was a fear that some patients might reject their proposed medication. RCs were seen as complementary to mental health services [34].

Those observations were echoed by Crowther et al [35]. Developing new relationships with users led to valuing coproduction, changed perceptions of users, and increased job satisfaction. The authors also concluded that “a certain distance is needed in the relationship between the Recovery College and its host organisation if a genuine cultural alternative is to be created.” On the other hand, separate organizations might mean lower staff awareness about the RC and less potential impact on formal mental health services [34]. These are interesting propositions to compare with benefits and obstacles of NHS Trust affiliation (access to professional support and higher visibility vs lack of support due to competing priorities and limited autonomy due to bureaucratic rules) [16].

The Stockholm RC was “fully integrated” into the psychiatry organization and not only affiliated. The same challenges were identified, though: lack of management support and dedicated resources. Sagatun, on the other hand, is placed in the community and governed and run by persons with lived experience. It has been claimed that RCs working with local community mental health teams contributing with new ideas for activities could be a source of inspiration. RCs were able to draw on resources and the expertise of those teams [16]. Although Sagatun was known and appreciated by local mental health teams, these did not participate in RC activities. Crowther et al [35] found that a “third sector” RC expands its outreach in the community and has a positive influence on public attitudes to mental health. That could possibly be the case in the Norwegian setting but was not assessed in this study.

If closer collaboration with users, including participating in RCs, is a key to developing more user-centered practices in mental health organizations, how could mental health professionals' willingness to more actively collaborate with users and engage in RCs be promoted? One of the authors discussed how social psychology theories might guide such an effort [36]. Showing the benefits of user involvement to professionals and how those resonate with their professional beliefs and goals will have a positive effect on their attitudes toward coproduction. Engaging professional champions to lead coproduction will create social pressure and establish norms that will attract other professionals to participate.

This study has both strengths and weaknesses. Among strengths are that the authors have a decade of experience of the 2 sites with unlimited access and ongoing communication with RC staff. They also have detailed knowledge of the organizational context of the cases. An additional strength is the possibility to systematically compare RCs with differing organizational characteristics, situated in 2 countries: one is embedded in a psychiatry organization, which is part of a regional public health system, and the other is a freestanding charity governed by users with no formal contacts with public mental health services. A weakness is the limited size of the study in terms of informants. Being of qualitative nature, the results of this research have to be interpreted with caution and might have limited generalizability outside the studied regions.

This multiple case study compared an RC embedded in a psychiatry organization with a freestanding, user-led RC. The RC contents showed only small differences; the embedded included more practical advice of formal services available, while the freestanding focused on the practice of social skills. The benefits experienced by participants matched those reported in the literature, including less experience of stigma, improved coping skills, and a reduction of service utilization.

Layperson SUEs employed in the psychiatry organization focused on safeguarding patient rights and improving patient involvement but were anxious to emphasize the value of evidence-based psychiatric care. Mental health staff had limited knowledge of the RCs on both sites. They still had interesting reflections on their workplaces in relation to recovery. In the psychiatry organization hosting the embedded RC, professional staff highlighted the “diagnosis-orientation” of their workplace and thought that a wider exposure to the RC could support a reorientation and promote a more patient-centered care culture and, as a result, improve collaboration between clinical units. PSWs (laypersons with user experience) engaged at the freestanding center were said by mental health professionals to be “go-betweens” and “bridge builders” to mental health services.

Overall, the differences between the RCs in views both by RC participants and mental health staff were small. The assumption at the outset of the study was that higher awareness and patient-professional collaboration would be found in the embedded RC but was not supported by these findings. Still, recent literature gives some advice on how to build on the insights from this study. A more recovery-oriented environment in mental health is promoted by engaged mental health professionals who have witnessed the benefits by actively participating in RCs. They realize the importance of collaboration with users and can act as champions to trigger change in clinical practice and care culture. The organizational form of the RC and its proximity to formal mental health services seem to be of less significance. People are more important than places.