Comparison of Nursing and Peer Support Worker Support in Crisis Plans for People with Mental Disorders in Japan: A Pilot Study Using Content Analysis Within a Realist Evaluation Framework

Abstract

This pilot exploratory qualitative study used the Modified Grounded Theory Approach (M-GTA) and the Realist Evaluation (RE) framework to clarify the interrelationships between nursing support and Peer Support (PS) in the creation and utilization of Crisis Plans (CPs) in Japan. Findings revealed that nursing support, constrained by psychiatric ward institutions, struggled with internal conflict while seeking to bridge risk and recovery. In contrast, PS support, rooted in lived experience, reframes crisis and risk, actively fostering patient self-determination and growth, and transforming. We propose an exploratory Context–Mechanism–Outcome (CMO) model that maps the distinctive contributions and interactions of nursing and PS support. This model provides provisional insights to inform the development of more robust, recovery-oriented CP support systems.

1. Introduction

In Japan, the number of psychiatric beds per 100,000 population significantly exceeds that of Western countries, with an exceptionally long average length of stay of approximately 280 days [1]. This is partly due to an estimated 70,000 “social admissions” of patients who are compelled to be hospitalized due to insufficient community support [2]. This situation impedes patient autonomy and leads to violations of dignity. Furthermore, the frequent use of physical restraint and seclusion in clinical settings has been flagged as a human rights issue [3]. Forced treatments such as involuntary admission, chemical sedation, physical restraint, and protective seclusion are often employed to address the acute deterioration of mental symptoms or imminent life crises. However, such coercive measures are known to inflict substantial harm, including physical and psychological distress, Post-Traumatic Stress Disorder (PTSD), deterioration of the therapeutic relationship, and increased self-stigma [3].

Against this backdrop, interventions to prevent involuntary hospitalization and forced measures in Japanese psychiatry have gained international attention. The Crisis Plan (CP), which encourages patients to be actively involved in their own care, has become a focus. A CP is a document planned in advance that expresses the individual’s preferred or non-preferred courses of action, anticipating impaired decision-making due to the exacerbation of mental symptoms. The origin of the CP lies in the Psychiatric Advanced Directives (PADs) developed in the US during the 1980s to protect patient rights [4]. While PADs are prevalent in the US, Joint Crisis Plans (JCPs) emerged primarily in the UK from the 2000s onwards, driven by the global paradigm shift towards abolishing coercive treatment and promoting community-based care [5]. JCPs are CPs developed collaboratively by mental health professionals and service users, and have been implemented as standard care for crisis intervention to avoid involuntary admission and coercive measures like physical restraint [6]. However, results from Randomized Controlled Trials (RCTs) indicated that the use of JCPs did not significantly reduce coercive treatment or enhance self-affirmation, suggesting that incorporating the patient’s preferences and hopes is essential for JCPs to be clinically effective [6].

In Japan, CP-J, which follows the UK’s JCP model, is being disseminated. Additionally, CPs within the Wellness Recovery Action Plan (WRAP) self-help program, which descends from PADs, are mainly utilized within peer support contexts [7]. However, there is criticism that the CP-J prevalent in Japan heavily emphasizes crisis intervention strategies and treatment planning by medical professionals, potentially failing to fully reflect the original intent of person-centeredness and human rights advocacy. Specific challenges include the tendency towards standardized care that overlooks the patient’s individuality and dynamic factors, a significant divergence between professionals and patients in the perception of risk, the underestimation of patient autonomy, and the resultant potential for limiting patient liberty and rights.

The importance of patient self-determination is globally emphasized, making the resolution of these issues urgent. Prior research indicates that barriers to decision-making include support systems that prioritize risk management, inadequate inter-professional collaboration, and insufficient trust-building between patients and professionals [8]. Francis et al. organized these barriers into a three-tiered structure—“organizational barriers,” “process barriers,” and “relational barriers”—suggesting that “fear” and “distrust” underlie them [8]. Furthermore, it has been reported that “crisis” and “risk” are fundamentally distinct concepts, with inherent differences in how service users and professionals perceive them [9]. This structural and conceptual misalignment is considered a factor in CP diverging from its original intent.

In this context, the involvement of Peer Support Workers (PSs) has garnered attention. PS are individuals who leverage their lived experience with disability or illness to provide support for others in similar situations, often employed by organizations to offer consultation and advice from a peer perspective. PS involvement is reported to include not only direct, peer-to-peer support for inpatients but also skilled support, such as mediation with professionals and facilitating relationship building with family members.

International studies suggest that PAD support involving PS, without intervention from clinical professionals, for individuals with severe mental illnesses such as schizophrenia or mood disorders, may lead to reduced involuntary admissions, mitigation of subjective symptoms, and enhanced empowerment [10]. This fosters the expectation of promoting Supported Decision-Making rather than acting as a proxy.

However, a survey by the Ministry of Health, Labour and Welfare indicated that the employment rate of PS in psychiatric medical institutions in Japan is low, at about 7% [11]. Challenges remain in establishing an environment where PS can be effective, including acceptance within professional-centric workplaces, the workplace atmosphere/culture, and training systems. Among inpatients in Japanese psychiatric hospitals, approximately half of long-term patients (admitted for 20 years or more) are still under involuntary commitment (e.g., compulsory or medical protection admission) [2]. Alongside Japan’s prominent issue of long-term hospitalization, safeguarding patient rights and supporting decision-making is a pressing matter.

While evidence supporting the effectiveness of PS-led CP support is accumulating, research comparing and examining the distinct support approaches of nurses and PS, who possess different professional expertise, and their interaction during CP creation and utilization is extremely limited. Specifically, empirical findings regarding the actual status and challenges of CP support collaboratively provided by nurses and PS are scarce in Japan.

Therefore, this study aims to clarify the characteristics and interrelationships of support provided by nurses and Peer Support Workers in the creation and utilization of Crisis Plans (CPs) in Japan, and to obtain exploratory theoretical insights for establishing a foundation for person-led support.

2. Materials and Methods

2.1. Study Design

This study employed an exploratory qualitative design that combined the Modified Grounded Theory Approach (M-GTA) with the CMO framework of Realist Evaluation (RE). This design was chosen because M-GTA provides the rigor for inductive, contextually grounded analysis essential for qualitative research, while the RE framework offers a structure for abductive reasoning to explain how and why observed support processes led to particular outcomes in a complex system. This report adhered as closely as possible to the Consolidated Criteria for Reporting Qualitative Research (COREQ) and the Realist Evaluation reporting guidelines (RAMESES II).

RE is a theory-driven evaluation method based on social constructivism [12]. It aims to clarify “what works, for whom, in what circumstances (Context), through what mechanisms (Mechanism), to produce what outcomes (Outcome)” when evaluating interventions and social programs implemented in complex contexts. Recently, it has been applied to evaluate educational programs and services in the healthcare domain, including nursing.

In this study, the interview data were analyzed using M-GTA based on the RE framework of “C + M = O (Context + Mechanism = Outcome)”. Analysis based on C + M = O incorporates deductive approaches using pre-existing theoretical frameworks, inductive reasoning to find new meaning in data, and abductive reasoning to theoretically reconstruct the underlying causal mechanisms behind observed phenomena. Abduction “recontextualizes” knowledge formally structured by deductive and inductive approaches, linking it to alternative interpretations to generate new ideas. The retrospective inference process that synthesizes these approaches is called retrodiction, a characteristic methodology of RE [13]. Given the novelty and scarcity of the research topic, the integration of M-GTA and RE’s C + M = O framework was employed to explain the causal relationships among the deep concepts and theories generated by M-GTA, thereby yielding practical implications [14].

Furthermore, this study compared the support environments of a general hospital psychiatric unit, which focuses on short-term acute care predominantly supported by professionals, with a specialized psychiatric hospital that includes PS in its multidisciplinary treatment programs, offers discharge support for long-term inpatients, and provides consultation support. By comparing the support content of professional nurses and non-professional PS, the study aimed to examine the roles and influence of PS, the necessity of shifting from treatment-centric support to one more focused on social reintegration, the utility of peer support, and the optimal form of collaboration with professionals to promote these goals.

2.2. Realist Evaluation Framework (C + M = O)

The study adopted the standard inferential procedure of Realist Evaluation (Pawson & Tilly, 1997) [12], the CMO framework, to identify what outcomes (“O”) an intervention (CP support) generates, through what mechanisms (“M”), and in what contexts (“C”) [13]. The categories generated by M-GTA were interpreted as components of this CMO framework, focusing on “C” which represents the characteristics of Japanese mental healthcare, particularly as “M” and “O”, in an attempt to generate middle-range theories explaining the success of the support.

2.3. Participants

The study involved a total of four participants: two nurses and two PS who were engaged in the creation and utilization of CP with patients in psychiatric wards.

The PS were formally employed as part-time hospital staff. A prerequisite for PS was a personal recovery experience from mental illness, and they had completed a designated public training program. Their selection was conducted by a multidisciplinary team based on years of recovery experience, aptitude as a supporter, and completion of training.

Researchers initially recruited accessible supporters (nurses and PS) through convenience sampling, followed by utilizing snowball sampling to recruit other participants. This method aimed to secure participants with diverse perspectives who had practical experience in CP support.

2.4. Data Collection

Semi-structured interviews were conducted with the four participants regarding their collaboration and support during the CP creation and utilization process with patients. Participants primarily shared their experiences based on the following core questions: “What is your specific role and core belief in CP creation and utilization?” “Could you share an episode where you found collaboration with nurses/PS particularly challenging or successful?” “What criteria or perspective do you use when judging a patient’s ‘crisis’ or ‘risk’?”.

2.5. Data Analysis

Integration of M-GTA and Realist Evaluation Framework:

The analysis followed the M-GTA procedure (Kinoshita, 2020) [14] to generate concepts and categories inductively from the participants’ accounts, emphasizing constant comparative analysis. To ensure analytical transparency and empirical depth, the following steps were performed:

1. Inductive Phase (M-GTA): Concepts and categories were generated, capturing the processes and contextual factors of CP support directly from the raw data excerpts. This stage was primarily concerned with “what is happening?”.

2. Conceptual Mapping Phase (CMO Construction—Abduction/Retrodiction): The inductively generated categories were systematically mapped onto the core components of the RE framework (Pawson & Tilly, 1997) [12]. The decision-making process for designating a category as Context (C), Mechanism (M), or Outcome (O) was based on its functional role within the causal chain inferred from the data and rigorously documented in analytical memos.

Context (C): Categories describing the institutional setting, regulatory factors, and clinical conditions surrounding the support were designated as Context.

Mechanism (M): Categories representing the distinctive reasons why the support worked (e.g., shared experience, risk reframing, clinical boundary setting) were designated as Mechanism. These were derived from the participants’ perceived actions and internal thoughts.

Outcome (O): Categories defining the resulting changes in the patient and the care environment (e.g., self-determination, relational trust, prevention of coercive intervention) were designated as Outcome.

3. Documentation of Transparency (Empirical Grounding): To directly address the need for empirical depth, detailed analytical memos (Bun-seki worksheet) were maintained throughout the process. Crucially, in Section 3, we provide extended, illustrative quotations (quotations) directly derived from the data to demonstrate the inductive emergence of the categories and to justify the final CMO mapping. Supplementary Material S1 provides further excerpts from the analytical memos detailing the transition from raw codes to final CMO categories, ensuring a traceable audit trail.

2.6. Mapping M-GTA Categories to the CMO Framework

Following the development of core categories using M-GTA, we employed a systematic mapping process to translate these qualitative findings into the components of the RE framework (C, M, O). We then functionally classified the M-GTA categories based on their roles.

As a crucial final step, we integrated C, M, and O through retroduction and abduction. Retroduction is the process of reasoning backward from observed outcomes (O) in the data to potential mechanisms (M) and necessary context (C). For example, the outcome (reduced anxiety in the son and first steps toward self-care in the mother) was retracted to the mechanism (PS’s advice to the mother led to activation of autonomy and clarification of needs), which was triggered by the context (family coping with repeated readmissions). Using abduction, we selected the most plausible and empirically grounded causal chain—the CMO relationship—that best explains “how and why” the intervention worked (e.g., Participant C’s statement: “The top priority is for the mother herself not to become ill”, which shifted focus from patient risk to family resources).

2.7. Analytical Transparency and Rigor

As part of the coding process, 122 codes were extracted and managed in Microsoft Excel. Two researchers (an expert in psychiatric nursing and an expert in public health) independently performed the initial coding, followed by a discussion of the results. Any disagreement in opinions was resolved through consensus-building discussion involving a third researcher (an expert in psychiatry and nursing research).

Researcher Reflexivity: The lead author has extensive clinical experience as a psychiatric nurse, recognizing the potential for this experience to introduce bias into the interpretation of data concerning the “medical model” or “risk management”. Therefore, during data interpretation, a critical discussion with co-authors was frequently held, deliberately striving to prioritize the service user’s perspective.

Translation and Validation: After the interview data (Japanese) were translated into English by a bilingual researcher, an independent external translation expert conducted content validation to ensure that the accurate nuances of the original meaning were preserved.

3. Results

3.1. Participant Attributes and Characteristics and CP Support Methods (Table 1)

Table 1 presents the attributes and characteristics of the research participants. The total of four participants included two nurses (Mr./Ms. A and B, both in their 30s) working in the psychiatric ward of General Hospital, and two PS (Mr./Ms. C in their 40s and D in their 20s) working at Psychiatric Hospital. Mr./Ms. A and B had over 10 years of experience, promoted CP creation and utilization as discharge promotion coordinators within the ward, and actively engaged in the process themselves. Mr./Ms. C and D had 1–2 years of PS experience and provided CP support to service users who had requested the WRAP program or were referred by professionals. All participants agreed to cooperate with the research.

Table 1. Subject Attributes and Characteristics and CP Support Methods.

Participant	A	B	C	D
Age	30s	30s	40s	20s
Occupation and Years of Experience	Nurse, over 10 years	Nurse, over 10 years	Peer staff worker, 2 years; Psychiatric social worker, 1 year	Peer staff worker, 1 year
Affiliated Facility	General hospital psychiatric ward	General hospital psychiatric ward	Psychiatric hospital	Psychiatric hospital
Responsibilities and Roles	Ward nursing and discharge support	Ward nursing and discharge support	Daycare program, CP support	Daycare program, CP support
CP Style	CP-J	CP-J	WRAP	WRAP
Timing of CP Creation	Conducted by the assigned nurse or nurse in charge after obtaining permission from the attending physician	Conducted by the assigned nurse or nurse in charge after obtaining permission from the attending physician	Conducted by a multi-professional team and the individual after obtaining permission from the attending physician and based on their request	Conducted by a multi-professional team and the individual after obtaining permission from the attending physician and based on their request

Table 1. Subject Attributes and Characteristics and CP Support Methods.

Participant	A	B	C	D
Age	30s	30s	40s	20s
Occupation and Years of Experience	Nurse, over 10 years	Nurse, over 10 years	Peer staff worker, 2 years; Psychiatric social worker, 1 year	Peer staff worker, 1 year
Affiliated Facility	General hospital psychiatric ward	General hospital psychiatric ward	Psychiatric hospital	Psychiatric hospital
Responsibilities and Roles	Ward nursing and discharge support	Ward nursing and discharge support	Daycare program, CP support	Daycare program, CP support
CP Style	CP-J	CP-J	WRAP	WRAP
Timing of CP Creation	Conducted by the assigned nurse or nurse in charge after obtaining permission from the attending physician	Conducted by the assigned nurse or nurse in charge after obtaining permission from the attending physician	Conducted by a multi-professional team and the individual after obtaining permission from the attending physician and based on their request	Conducted by a multi-professional team and the individual after obtaining permission from the attending physician and based on their request

All four participants provided CP creation and utilization support to inpatients. Mr./Ms. C also had one year of experience as a Psychiatric Social Worker (PSW). Mr./Ms. A and B, as part of the discharge promotion team in the acute care ward, individually supported and promoted CP creation for all inpatients and professionals. Mr./Ms. C and D provided CP.

Support in collaboration with professionals upon request from patients or multidisciplinary staff. In both facilities, the CP was created through several dialogues during the pre-discharge phase and finalized through cycles of trial and revision.

3.2. Structure of Major Categories and Sub-Categories (Table 2)

Table 2 systematically compares the structure of CP support provided by nurses and PS across three dimensions based on the “CMO” framework. Nursing support, under the “C” of [Treatment responsibility and Systemic Conflict with Recovery Model] context, employed the “M” of [Bridging Risk and Recovery Implementation of CP with concurrent Introspection] mechanism, leading to the “O” of [Clinical Recovery and Reflective restructuring of care philosophy]. Conversely, PS support, based on the “C” of [Resource Mobilization of Lived Experience and Questioning of Professional Perspectives] context, employed the “M” of [Reconstruction of Meaning and Activation of Self-Determination] mechanism, resulting in the distinct “O” of [Empowerment of Agency and Structural Transformation of the care system].

Table 2. Experiences of Nurses and PS in Providing CP Support.

Context	Mechanism	Outcome
Background and Context for Providing CP Support	Mechanisms that Produce Results	Results and Achievements
Nurses’ CP Support Experience
Treatment responsibility and Systemic Conflict with Recovery Model	Bridging Risk and Recovery Implementation of CP with concurrent Introspection	Clinical Recovery and Reflective restructuring of care philosophy
Acute Care Priority and Mandate for Safety Assurance	CP used as a Therapeutic Tool Promoting patient insight through verbalization of symptoms	Improved Insight and Therapeutic Alliance
Tendency to Equate Risk and Crisis influenced by the medical model	Balancing Safety Management with seeking continuity in Discharge Support	Emergence of Risk Differentiation Awareness and specific discharge planning
PS’ CP Support Experience
Resource Mobilization of Lived Experience and Questioning of professional perspectives	Reconstruction of Meaning and Activation of Self-Determination	Empowerment of Agency and Structural Transformation of the care system
Trust in the Value of the client’s self-expression	Presentation of Resourced Needs to the team	Clarification of Needs and Deepening of Nurses’ Engagement
Awareness of Rigidity in Support Structure and interprofessional dynamics	Team Readjustment questioning conventional support and establishing a collaborative environment	Team Empowerment and Recovery-Oriented System transformation

Table 2. Experiences of Nurses and PS in Providing CP Support.

Context	Mechanism	Outcome
Background and Context for Providing CP Support	Mechanisms that Produce Results	Results and Achievements
Nurses’ CP Support Experience
Treatment responsibility and Systemic Conflict with Recovery Model	Bridging Risk and Recovery Implementation of CP with concurrent Introspection	Clinical Recovery and Reflective restructuring of care philosophy
Acute Care Priority and Mandate for Safety Assurance	CP used as a Therapeutic Tool Promoting patient insight through verbalization of symptoms	Improved Insight and Therapeutic Alliance
Tendency to Equate Risk and Crisis influenced by the medical model	Balancing Safety Management with seeking continuity in Discharge Support	Emergence of Risk Differentiation Awareness and specific discharge planning
PS’ CP Support Experience
Resource Mobilization of Lived Experience and Questioning of professional perspectives	Reconstruction of Meaning and Activation of Self-Determination	Empowerment of Agency and Structural Transformation of the care system
Trust in the Value of the client’s self-expression	Presentation of Resourced Needs to the team	Clarification of Needs and Deepening of Nurses’ Engagement
Awareness of Rigidity in Support Structure and interprofessional dynamics	Team Readjustment questioning conventional support and establishing a collaborative environment	Team Empowerment and Recovery-Oriented System transformation

3.3. “C + M = O” Structure of CP Creation and Utilization Support Experiences in Nurses

The specific CMO configurations derived from the nurses’ CP support experience are shown.

CMO Chain 1: Under the “C” of [Treatment responsibility and Systemic Conflict with Recovery Model], the “M” of [Bridging Risk and Recovery Implementation of CP with concurrent Introspection] operated.

C1:

“Is it a characteristic of the hospital, or something? Patients get transferred really quickly, and then they’re just gone.” “For example, when we talk about them wanting to use their hobbies or skills to live or work, and then it comes down to, ‘well, what about your illness first?’ Then, it feels like there are limits to what we can try or do during the hospital stay, haha.” (Participant B)

“Yes, well, our ward does have a lot of readmissions. Even with home-visit nursing services and even with this notebook (CP), the readmission rate is quite high, around 70%.” (Participant A)

M1:

“So, to make it clearer what they should think about specifically, how they should ask for support, or who would help them, we gave them the ‘Life Notebook’ (CP) and had them write things down. I think that way, all the other staff members started paying attention and consciously doing it too.” (Participant B)

“Well, one successful case was being able to review things, realizing that enduring distress for too long is tough, so we encouraged them to use their PRN (as-needed) medication earlier. It was a nursing intention and desire that they find a coping method that suits them, not just relying on medication. (Participant A)

CMO Chain 2: Under the “C” of [Acute Care Priority and Mandate for Safety Assurance], the “M” of [CP used as a Therapeutic Tool Promoting patient insight through verbalization of symptoms] operated, resulting in the “O” of [Improved Insight and Therapeutic Alliance].

C2:

“Yes, well, gradually after the ECT(Electro Convulsive Therapy) finishes, they have their own life, and some people truly fall to the bottom (of their level? memory?) and then recover with the ECT electricity.” (Participant A)

M2:

“However, from the nurses’ perspective, regarding those patients who don’t express much—we might use the Life Note as an approach, meaning using it as a communication tool.” (Participant A)

CMO Chain 3: The “C” of [Tendency to Equate Risk and Crisis influenced by the medical model] saw the “M” of [Balancing Safety Management with seeking continuity in Discharge Support] operate, resulting in the “O” of [Emergence of Risk Differentiation Awareness and specific discharge planning]. Discrepancies in perception existed between patients, who sometimes approached CP as merely filling out a form, and professionals, who felt compelled to complete the process. Consequently, engagement with CP led nurses to re-recognize the necessity of support focused on post-discharge life, allowing them to discuss the crisis and its countermeasures even when patients were reluctant to accept their symptoms, and to focus on the patient’s personal goals.

C3:

“The CP process is difficult because while I want patients to start writing freely, the staff often treats it as a mandatory task they ‘have to do.” (Participant B)

M3:

“We utilize CP in the Discharge Support Conference to clarify how and from whom patients should seek support after discharge, making the planning concrete.” (Participant A)

O3:

“Calling the son back from the facility just because they’re lonely is too simplistic. Shouldn’t we explore alternatives like a cooking class, given their hobby?” (Participant B)

3.4. “C + M = O” Structure of CP Creation and Utilization Support Experiences in PS (Table 2)

CMO Chain 1: Under the “C” of [Resource Mobilization of Lived Experience and Questioning of professional perspectives], the “M” of [Reconstruction of Meaning and Activation of Self-Determination] operated, leading to the “O” of [Empowerment of Agency and Structural Transformation of the care system]. PS was motivated by the painful realization of societal issues, such as unilateral job transfers despite requests for reasonable accommodation. This motivated them to value “diligently accepting the words patients convey” and “thoughtfully considering things together” to prioritize the patient’s will.

C1:

“In my truly younger days, there were times when my mental symptoms caused intense excitement, so at those times, I really thought it’s better to put me in seclusion than to hurt someone—a preference I included in my CP… Patients often cannot speak to the doctor or simply follow the nurses’ instructions.” (Participant C)

M1:

“When someone experiences things like that, the person who spirals down inevitably develops a kind of self-stigma themselves.” “Because I am exposing myself in a way, things that were hard to say became easier to talk about, and there was a sense of security just from handing it over, knowing that they understand.” (Participant C) “By exposing myself through the CP, self-disclosure became easier…” and merely handing over the document brought a sense of security from being understood.” (Participant D)

CMO Chain 2: Under the “C” of [Trust in the Value of the client’s self-expression], the “M” of [Presentation of Resourced Needs to the team] operated, leading to the “O” of [Clarification of Needs and Deepening of Nurses’ Engagement]. Peer Staff (PS) were motivated by the painful realization of societal issues, such as the unilateral job transfers despite requests for reasonable accommodation. This motivated them to value “diligently accepting the words patients convey” and “thoughtfully considering things together” to prioritize the patient’s will.

C2:

“After working there for many years, when the organization was subject to a national policy requiring disclosure of employee disability status, I simply shared information about my disability—when there were no issues with my performance—I was suddenly transferred to duties different from those I had performed previously. Huh!? [Seriously?] I felt that was not accommodation… it was unilateral accommodation.” (Participant C)

M2:

“Drawing on my own experience with shared communication uniqueness, I acted as an intermediary during a care conference for a patient with a unique profile, saying, ‘Isn’t this unique world view actually about this specific kind of experience?’ which genuinely pleased the social worker… I entered the discussion saying, ‘This tone of voice is just normal for this person… It was something like advocacy.” (Participant C)

O3:

“After creating the Crisis Plan (CP) with patients, some have actually been discharged, and when the Crisis Plan was activated, the crisis was resolved. Those individuals often come back and show their faces when they visit the hospital. At that time, I truly feel it was amazing, and I think, ‘Ah, that was good.’” (Participant C) “I have felt that the people around me take my words, they become very considerate, and they speak, valuing each and every word. It feels like the people around me pay a little attention to my words, and they pause for a moment.” (Participant D)

CMO Chain 3: In the context of the “C” of [Awareness of Rigidity in Support Structure and interprofessional dynamics], the “M” of [Team Readjustment questioning conventional support and establishing a collaborative environment] was observed, resulting in the “O” of [Team Empowerment and Recovery-Oriented System transformation]. In supporting a patient who resisted independence due to dependence on family, PS Mr./Ms. C acted as a mediator, explaining the patient’s unique perspective to the team and family. PS subsequently facilitated a change in the support structure by boldly encouraging the family to clearly communicate the need for the patient to move to a group home due to the family’s fear.

C3:

They call it ‘team medicine,’ and they are supposed to look at things from various angles to come up with the best or a better answer, but each person asserts their own views, and I don’t know what’s what.” (Participant C)

M3:

“From my own experience: if you clearly state that something truly unforgivable is unforgivable, the other person will ultimately understand… Therefore, Mother, please convey that message to him very strongly and very clearly.” (Participant C)

O3:

“The mother actually told him, and the person (the patient) was momentarily crestfallen but then decided, ‘I’ll go to the group home.’… After that, he really made up his mind and went to the group home, and today his parents came and showed me a picture, and he looked truly happy.” “I really felt that a person who had been struggling so much in the hospital could change like that. It left a strong impression on me.” (Participant C)

4. Discussion

4.1. Positioning of Research Findings and Study Characteristics

This is an exploratory qualitative pilot study with a small sample size of only four participants (two nurses and two PS), aiming to construct an initial theoretical model of CP support within the specific context of a Japanese psychiatric ward. Consequently, these findings should be cautiously positioned as “Conceptual Implications” or “Empirical Hypotheses” to inform future large-scale realist validation studies, and cannot be broadly generalized to other cultural spheres or healthcare systems due to the characteristics of this small-scale design.

This qualitative study identified the core CMO relationships to inform future validation, providing an explanatory framework for CP support provided by nurses and PSs.

We acknowledge that a primary limitation of this study is the small sample size (n = 4), which limits the strength of the claim to “conceptual saturation”. However, the study’s unique methodology, which integrates M-GTA for deep, inductive grounding with the RE framework for structured, abductive explanation, allowed us to derive a theoretical structure from limited data. By restructuring the findings not merely as descriptions but as a theoretical model based on the interconnectedness of the intervention’s context (C) and mechanism (M), the CMO framework enabled the derivation of theoretical depth and practical implications from small-scale data.

4.2. Systematic Comparison of Roles Between Nurses and Peer Support Workers (PSs)

4.2.1. Support by Nurses: Internal Conflict and the Sprout of Structural Transformation

Our results indicate that nursing support for CP is characterized by the internal conflict of “Bridging Risk and Recovery,” which arises between the institutional constraint of treatment prioritization in psychiatric wards and the emerging ethical view of patient-led recovery. Under the context of treatment responsibility and institutional risk avoidance, nurses underwent a reflective process that attempted to reintroduce the perspective of Quality of Life (QOL) and human rights advocacy into a “safety-first” support view. This process is interpreted as the sprout of structural transformation, suggesting a restructuring of consciousness from conventional medical-model risk management to a multifaceted discrimination of risk that reflects the patient’s life and values. This provides an important counter-evidence to prior research that pointed to the rigidity of the support perspective among medical professionals [15].

4.2.2. The Role of PS and Dynamic Team Adjustment Through “Restructuring of Meaning”

In contrast, PS, based on a stance of re-examination rooted in their recovery experience, restructured the meaning of crisis and risk as a “challenge to overcome,” promoting the patient’s self-determination and growth. A particularly critical finding is that PS also functioned as a “dynamic adjuster”, dispersing excessive treatment responsibility from professionals and integrating a recovery-oriented perspective into the entire team through this restructuring of meaning. This mechanism suggests that PS is not merely a supporter but acts as a “catalyst” that exposes differences in the ethical foundations within multidisciplinary collaboration, thereby facilitating dialogue and the adjustment of the collaborative environment.

4.3. Critical Discussion on the Application and Limits of “Positive Risk-Taking”

The support provided by PS aligns with the concept of Positive Risk-Taking (PRT), which is based on the service user’s recovery experience and restructures risk as an “object to be overcome” [16]. However, based on the findings of this study, the practice of PRT must be critically and cautiously discussed, as it involves the following ethical and practical limitations and challenges.

First, there is a potential conflict with medical safety. Particularly for patients in the severe acute phase, the PS’s restructuring of risk inherently carries the danger of directly conflicting with the patient’s life safety or high-urgency medical interventions. While PRT aims to maximize the patient’s self-determination, it can lead to severe dilemmas between the conflicting ethical principles of the duty to protect and the respect for autonomy.

Second, the ambiguity of the boundary between professional and PS responsibilities. To safely promote PRT, the PS’s role must be clearly demarcated from the legal and professional treatment responsibilities of medical professionals and the patient’s own right to self-determination. The findings of this study suggest the danger of this boundary becoming ambiguous, highlighting the critical necessity of a clear Team Alignment regarding the division of roles within the multidisciplinary team and the accountability for legal responsibility in the event of a risk.

Therefore, to implement PRT safely and ethically, it is essential that professionals and PS establish a common ethical foundation regarding crisis and risk, and a clear, prior consensus on role division in the event of a risk. These findings emphasize that PRT is not unconditionally recommended, but that the adjustment of the collaborative environment to overcome the aforementioned challenges is critically important.

5. Conclusions

These findings are limited to the exploratory stage of theory generation derived from a small number of participants (N = 4), and broad generalization is not possible. The results provide concrete hypotheses for future large-scale CMO validation studies.

Based on this study, the following exploratory CMO hypotheses are proposed:

Establishing an Ethical Foundation: If the “M” is to conduct joint training for nurses and PS focusing on the “ethical limits of active risk-taking,” the “O” will be the establishment of a common ethical foundation for risk discrimination between the two professions.

Early Structural Transformation: If the “M” is to formally include PS in multidisciplinary team meetings from the early stage of the patient’s admission, the “O” will be the promotion of dialogue on “risk discrimination,” leading to reduced stigma and enhanced patient self-determination.

Safe Application of Risk-Taking: Under the “C” constraint of a closed psychiatric ward, if the “M” is the cautious application of risk-taking by PS under appropriate risk discrimination by medical professionals and a collaborative monitoring system with clear responsibilities, the “O” will be the coexistence of empowerment and medical safety.

These findings provide a theoretical foundation for the shift toward a recovery-oriented system and encourage future validation studies adhering to RAMESES II [17].