Cross-cultural de-centring of Recovery College tools: a 20-country Delphi study

Cross-cultural de-centring of Recovery College tools: a 20-country Delphi study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Cross-cultural de-centring of Recovery College tools: a 20-country Delphi study Yasuhiro Kotera, Sara Vilar-Lluch, Dagmar Narusson, Ramona Hiltensperger, and 17 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9614758/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Recovery Colleges are expanding internationally (264 identified in 31 countries across six continents), yet key tools for defining and assessing them were developed in high-income Anglophone contexts. We conducted a two-round international Delphi study using a cross-cultural de-centring approach to examine the cross-cultural applicability of the RECOLLECT Change Model and RECOLLECT Fidelity Measure, and to develop globally applicable and interoperable versions of these tools. In Round 1, 54 panellists from 20 countries rated item importance and cultural difficulty and provided free-text feedback, analysed using corpus-informed linguistic methods. Core principles were broadly endorsed, but “coproduction”, “recovery”, “professionals” and “shifting the balance of power” were interpreted inconsistently, revealing embedded assumptions about equality, autonomy and open-ended participation. Round 1 findings informed wording refinements tested in Round 2 with 41 panellists from 15 countries. All 11 revised items reached consensus for understandability and cultural appropriateness, strengthening the tools while preserving core Recovery College values. Health sciences/Health care/Health services Health sciences/Health care/Patient education Introduction Recovery Colleges (RCs) are an innovative approach to mental health support that combines co-production and adult learning to promote personal recovery and social inclusion. 1,2 Established in England in 2009, RCs have since expanded internationally. A previous global review identified 221 RCs in 28 countries in 2022, 3 and we are aware of further growth to 264 RCs worldwide by 2026. RCs provide inclusive educational environments in which people with mental health problems, carers, staff, and people in the wider community (referred to as “students”) learn together, develop skills and build support networks. Through self-directed learning and shared expertise, RCs support fuller social participation, greater self-management and personal recovery. 4 Emerging evidence indicates that RCs can improve student outcomes, including self-esteem, quality of life and internalised stigma, 5 and benefit staff through increased motivation and skill development. 6 A large retrospective matched cohort study involving 1,435 students and 4,665 controls found reduced subsequent mental health service use and a £5,028 greater reduction in total costs per student at 12 months, equivalent to an estimated 8.4:1 return on investment. 7 Uncontrolled before-and-after analyses in the same evaluation also showed improved symptoms and functioning. 7 These economic findings compare favourably with a 2025 estimate of an 8:1 socio-economic return for preventive healthcare interventions. 8 However, operational guidance for RCs has only recently become more standardised through the Recovery Colleges Characterisation and Testing (RECOLLECT) programme (NIHR200605). Two widely-used outputs from RECOLLECT are the RECOLLECT Change Model (RCM) and the RECOLLECT Fidelity Measure (RFM). The RCM identifies four mechanisms through which RCs are proposed to work: empowering environment; shifting the balance of power; enabling different relationships; and facilitating personal growth. 9 The RFM is a 12-item tool assessing fidelity to core RC principles and best practice. 10 The 12 items comprise seven non-modifiable items, representing core principles expected to be present across Recovery Colleges, and five modifiable items, representing features that may appropriately vary by local context. Non-modifiable items include equality, adult learning, tailoring to the student, co-production, social connectedness, community focus and commitment to recovery. Modifiable items concern accessibility, location, distinctiveness of course content, strength-based orientation and progressiveness; for example, whether access is universal or targeted, whether the setting is Recovery College-specific or community-integrated, and whether course content is distinct from or integrated with clinical services (Supplementary File 1). Developed in England through an empirical programme of work, these tools drew on a literature review, consultation with 77 international experts, interviews with 10 Recovery College managers, refinement with 44 students, trainers and managers across 3 sites, and evaluation involving 60 professional stakeholders. The international growth of RCs raises an important question: whether tools developed in one cultural context can be applied appropriately across others. This concern is particularly relevant because most RC research has been conducted in England, where approximately one third of all RCs are based, 11 and in other high-income Global North settings, limiting understanding of how RC principles and language operate across more diverse cultural, linguistic and service systems. 3,12 Cultural fit matters. A meta-analysis of 76 studies involving 25,225 participants found that culturally adapted mental health interventions showed a moderately strong advantage over unadapted approaches (d=.45). 13 A direct-comparison meta-analysis also found that culturally adapted psychotherapy outperformed unadapted psychotherapy (d=0.32). 14 Together, these findings indicate that cultural adaptation is not a peripheral concern: the language, examples and assumptions embedded in mental health interventions and implementation tools may materially affect whether they are understood, accepted and effective. Because the RCM and RFM are intended to be used across diverse RC contexts, the key methodological challenge was not only translation or local adaptation, but cross-cultural interoperability: preserving shared conceptual meaning while allowing wording and examples to fit different cultural and service contexts. A consensus exercise was therefore chosen to integrate international stakeholder judgement on how the RCM mechanisms and RFM items were interpreted across contexts, where agreement or disagreement indicated culturally different understandings, and how the wording should be refined to improve cross-cultural interoperability. This study aimed to identify the level of cultural influence on the RCM mechanisms and RFM items, and to develop globally interoperable versions of these tools. Results A total of 54 panellists from 20 countries completed Round 1. Participants included 22 RC managers (40.7%), 10 RC researchers (18.5%), 9 RC peer trainers (16.7%), 7 RC non-peer trainers (13.0%), and 6 other RC staff (11.1%). In terms of RC experience, 13 panellists (24.1%) had 7+ years’ experience, 13 (24.1%) had 4–6 years, 22 (40.7%) had 1–3 years, and 6 (11.1%) had less than 1 year. Geographically, panellists were mainly from Western Europe (42.6%) and Asia (29.6%), with additional representation from Central Europe (11.1%), Oceania (5.6%), North America (5.6%), South America (3.7%), and Eastern Europe (1.9%). In Round 2, 41 panellists (75.9%) from 15 countries participated. Geographical representation remained strongest from Western Europe (46.3%) and Asia (26.8%), followed by Central Europe (9.8%), North America (7.3%), Oceania (4.9%), and South America (4.9%). Table 1 presents the panellist characteristics. Table 1. Panellist characteristics (n=54) Country Primary role Years of experience R2 Australia RC staff 7+ years ✓ Australia RC manager 1–3 years ✓ Brazil RC manager 1–3 years ✓ Brazil RC researcher 1–3 years ✓ Canada RC researcher 7+ years ✓ Canada RC manager 1–3 years ✓ Canada RC manager 4–6 years ✓ Czech Republic RC manager 4–6 years ✓ Czech Republic / Slovakia RC researcher 4–6 years ✓ Denmark RC manager 7+ years ✓ Denmark RC non-peer trainer 7+ years ✓ Denmark RC peer trainer 1–3 years England RC researcher 1–3 years ✓ England RC manager 7+ years ✓ England RC manager 1–3 years England RC manager 7+ years ✓ England RC manager 4–6 years ✓ England RC staff 4–6 years ✓ England RC manager 7+ years England RC peer trainer Less than 1 year ✓ England RC peer trainer 4–6 years ✓ Estonia RC researcher Less than 1 year Finland RC researcher 1–3 years ✓ Finland RC manager 4–6 years ✓ Finland RC non-peer trainer Less than 1 year ✓ France RC researcher 7+ years ✓ Germany RC researcher 1–3 years ✓ Germany RC manager 7+ years Germany RC non-peer trainer 1–3 years ✓ Hong Kong RC manager 4–6 years Hong Kong RC researcher 7+ years Italy RC manager 1–3 years ✓ Italy RC manager 7+ years ✓ Japan RC non-peer trainer 7+ years ✓ Japan RC non-peer trainer 4–6 years ✓ Japan RC peer trainer 1–3 years ✓ Japan RC researcher 1–3 years ✓ Japan RC manager 4–6 years ✓ Japan RC manager 1–3 years Japan RC non-peer trainer 1–3 years ✓ Japan RC non-peer trainer 4–6 years ✓ Netherlands RC researcher Less than 1 year ✓ New Zealand RC staff 1–3 years Spain RC manager 7+ years Sweden RC researcher 1–3 years ✓ Sweden RC manager Less than 1 year ✓ Sweden RC manager 1–3 years ✓ Switzerland RC manager 1–3 years Thailand RC peer trainer 1–3 years ✓ Thailand RC peer trainer 7+ years ✓ Thailand RC peer trainer Less than 1 year Thailand RC peer trainer 4–6 years Thailand RC peer trainer 4–6 years ✓ Thailand RC staff 1–3 years ✓ R2 = participated in the Round 2 survey. RC = Recovery College. RC non-peer trainer includes roles such as psychologist, nurse, social worker, or educator. Round 1 quantitative findings All four RCM mechanisms reached consensus for importance, with 92.6%–98.1% of panellists rating them in the 4–5 category after collapsing the five-point scale into three categories (Table 2). Six of the seven RFM non-modifiable items also reached consensus for importance (70.4%–100.0%). The exception was tailored to the student, which fell just below the threshold (37/54, 68.5%). Median importance ratings were uniformly high (median 4–5). By contrast, consensus on cultural difficulty was more limited. Crucially, no RCM mechanism reached consensus for low difficulty (1–2 category). Among the RFM non-modifiable items, only Learning, Social connectedness, and Commitment to recovery reached consensus for low difficulty, with 70.4%, 70.4%, and 75.9% of panellists respectively rating them in the 1–2 category. Median difficulty ratings ranged from 2 to 3 across items, indicating that achieving these standards involves significant cultural navigation. Table 2. Round 1 ratings for importance and cultural difficulty of RCM mechanisms and RFM non-modifiable items (N=54) Item Importance 4–5 n (%) Importance Median (IQR) Importance consensus met Difficulty 1–2 n (%) Difficulty Median (IQR) Difficulty consensus met RCM mechanisms Empowering environment 51 (94.4) 5 (4–5) Yes 33 (61.1) 2 (2–3) No Shifting the balance of power 50 (92.6) 5 (4–5) Yes 22 (40.7) 3 (2–4) No Enabling different relationships 52 (96.3) 5 (4–5) Yes 28 (51.9) 2 (2–3) No Facilitating personal growth 53 (98.1) 5 (5–5) Yes 34 (63.0) 2 (2–3) No RFM non-modifiable items 1. Valuing equality 54 (100.0) 5 (5–5) Yes 33 (61.1) 2 (2–3) No 2. Learning 48 (88.9) 5 (4–5) Yes 38 (70.4) 2 (2–3) Yes 3. Tailored to the student 37 (68.5) 4 (3–5) No 20 (37.0) 3 (2–3) No 4. Co-production of the RC 49 (90.7) 5 (4–5) Yes 27 (50.0) 2 (2–3) No 5. Social connectedness 50 (92.6) 5 (4–5) Yes 38 (70.4) 2 (2–3) Yes 6. Community focus 38 (70.4) 4 (3–5) Yes 24 (44.4) 3 (2–4) No 7. Commitment to recovery 51 (94.4) 5 (5–5) Yes 41 (75.9) 2 (1–2) Yes Consensus was predefined as ≥70% of responses falling within the same collapsed category. For importance, consensus was assessed in the 4–5 category. For difficulty, consensus was assessed in the 1–2 category. Medians and IQRs (Interquartile Ranges) are based on the original five-point scale. Round 1 qualitative findings informing revision To understand how panellists interpreted the items across settings and to identify wording that could improve cross-cultural applicability, free-text responses from Round 1 (totalling 9,541 words) were analysed using corpus-informed linguistic analysis. This analysis extracted ‘keywords’ (terms significantly salient in the focus dataset in comparison to a larger one of general English language) for each dataset of the free-text responses to identify recurrent terms and patterns of language use, and inductively identify themes. Examination of the keywords in context of use can identify ambiguities and points of tension in how panellists understood and applied the RC wording/terminology. Table 3 presents the main cross-cultural issues identified (Supplementary File 3 for more details). Table 3. Main cross-cultural issues identified in Round 1 free-text responses and corresponding wording revision Item Panellist feedback Revision made Empowering environment Lacked a clear equivalent in some contexts; meaning was inconsistent Emphasised strengths, autonomy, choice and control Shifting the balance of power Felt abstract, confrontational or hard to realise; cultural and structural barriers noted Reframed as equal partnership Enabling different relationships Unclear what relationships were meant and with whom Revised to focus on connecting and working together Facilitating personal growth Expectation of upward progress did not fit recovery in some contexts Recentred on recovery rather than growth Valuing equality Tension between equality and individual needs; “student” label queried Emphasised accessibility, equity and clearer role descriptors Learning Adult learning approaches were unfamiliar in some settings; terminology was too jargonistic Simplified and clarified in non-jargonistic terms Tailored to the student Interpreted variously as person-centred flexibility or cultural adaptation; resource constraints noted Clarified as adapting to student needs within local context Co-production of the RC Meaning unclear; panellists requested explanation or examples Described shared decision-making and contributor roles more explicitly Social connectedness Ambiguity between internal RC relationships and wider community links Distinguished connectedness within the RC and beyond Community focus Unclear what kind of community was meant; organisational barriers noted Made intended meaning of community more explicit Commitment to recovery Often understood as clinical rather than personal or social recovery Clarified intended meaning of recovery Free-text feedback on the five modifiable RFM items showed that the original binary Type 1/Type 2 response format did not adequately reflect all RC models. Panellists noted that some colleges combined both types within the same item, such as being partly community-based and partly located in a dedicated RC space, or offering course content both distinct from and integrated with clinical provision. Therefore, a third response option, “Both Type 1 and Type 2 (blended)”, was added for all modifiable items before Round 2. Supplementary File 4 presents the wording changes made for Round 2 in response to Round 1 input. Items that had already reached consensus as culturally not difficult (i.e., RFM 2 Learning, 5 Social connectedness, and 7 Commitment to recovery) were retained largely unchanged; where qualitative comments informed revision, these changes were minor and intended only to clarify wording or improve cross-cultural accessibility. Round 2 evaluation of revised wording In Round 2, all 11 revised items reached consensus for both understandability and cultural appropriateness (Table 4). For understandability, 73.2%–95.1% of panellists rated the revised wording in the 4–5 category, and for cultural appropriateness the corresponding range was 80.5%–95.1%. These findings supported the acceptability of the revised wording across settings. Table 4. Round 2 quantitative findings Revised item Understandable in context 4–5 n (%) Median (IQR) Consensus Culturally appropriate 4–5 n (%) Median (IQR) Consensus RCM mechanisms Empowering, autonomy-supporting environment 30 (73.2) 4 (3–4) Yes 34 (82.9) 4 (4–4) Yes Moving toward an equal partnership 39 (95.1) 5 (4–5) Yes 38 (92.7) 5 (4–5) Yes Connecting with peers and working together 37 (90.2) 5 (4–5) Yes 37 (90.2) 5 (4–5) Yes Facilitating recovery 32 (78.0) 4 (4–5) Yes 33 (80.5) 4 (4–5) Yes RFM non-modifiable items 1. Valuing equality 39 (95.1) 4 (4–5) Yes 37 (90.2) 5 (4–5) Yes 2. Learning 37 (90.2) 4 (4–5) Yes 36 (87.8) 5 (4–5) Yes 3. Tailored to the student 36 (87.8) 4 (4–5) Yes 36 (87.8) 4 (4–5) Yes 4. Co-production of the RC 33 (80.5) 4 (4–5) Yes 34 (82.9) 4 (4–5) Yes 5. Social connectedness 39 (95.1) 4 (4–5) Yes 39 (95.1) 5 (4–5) Yes 6. Community focus 39 (95.1) 4 (4–5) Yes 35 (85.4) 5 (4–5) Yes 7. Commitment to recovery 39 (95.1) 5 (4–5) Yes 39 (95.1) 5 (4–5) Yes Consensus was predefined as ≥70% of responses in the 4–5 category. Finalised cross-cultural version of the RCM and RFM items In Round 2, all items reached consensus as culturally appropriate. Qualitative comments (1,157 words) were limited and largely confirmatory. Three minor wording refinements were implemented by the research team (Supplementary File 5). The final post-Round 2 items are presented as the Global RCM Version 1 (Global RCM v1) and Global RFM Version 1 (Global RFM v1), alongside the original England-developed wording, in Supplementary File 6. Discussion This two-round international Delphi study used a cross-cultural de-centring approach, informed by corpus-based linguistic analysis of panellist feedback, to examine the cross-cultural applicability of the RCM and RFM. We found that cultural influence operated less at the level of whether RC principles were valued, and more at the level of how those principles were understood, worded and enacted. Core principles were broadly endorsed across countries, but concepts embedded in England-developed RC tools and wider high-income Anglophone recovery-oriented discourse, such as coproduction, shifting power and recovery itself, were not always directly applicable across settings without clarification. Cross-cultural de-centring helped make implicit assumptions explicit, distinguish personal and social recovery from clinical recovery, and reframe ideas such as tailoring and partnership in ways that were meaningful and feasible across diverse contexts. This is timely in light of current international emphasis on person-centred, rights-based and community-based mental health care. 15 A central finding was that the underlying RC principles travelled better than the original English wording used to express them. In Round 1, all four RCM mechanisms and six of seven non-modifiable RFM items reached consensus for importance, yet consensus on low cultural difficulty was much more limited. This pattern suggests that panellists across settings broadly endorsed the constructs, but not always the phrasing through which those constructs had been operationalised. In cross-cultural psychiatry terms, the challenge was less about conceptual rejection than about semantic portability and contextual intelligibility. 2 The study therefore supports a distinction between valuing a principle and recognising it in a linguistic form shaped by a particular cultural and service discourse. This distinction is especially important for recovery-oriented frameworks, where apparently familiar terms may carry different institutional, clinical and moral meanings across settings. 16 The study also makes visible how tools developed in high-income Anglophone settings can contain assumptions that are locally intelligible but not automatically universal. Terms such as “coproduction”, “shifting the balance of power” and “recovery” are well established within UK recovery-oriented policy and service reform discourse, and can therefore appear self-explanatory to researchers and practitioners working within that context. Internationally, however, these terms were not always interpreted consistently and sometimes carried unintended meanings. 17,18 For example, “shifting the balance of power” assumes a discourse of explicit power redistribution that may resonate in Anglophone service reform language, but may sound confrontational, abstract or structurally unclear elsewhere. 19 Likewise, although coproduction is now widely used in UK health policy and personalised care discourse, it does not travel as a universally shared term without explanation. 20 One interpretation is that this wording reflected assumptions common in some Anglophone service cultures, including relatively explicit egalitarian language, individual autonomy and tolerance of open-ended participation. 21,22 Across countries, panellists repeatedly sought clarification of meaning, role and relationship, suggesting that these assumptions were not consistently shared internationally. De-centring therefore did not replace the underlying RC principles; rather, it made their intended meaning more explicit and contextually interpretable for broader international use. This aligns with wider concerns in cross-cultural psychiatry about dominant systems mistaking local categories for universal ones. 23 The item revisions illustrate how this de-centring process worked in practice. “Empowering environment” became “Empowering, autonomy-supporting environment”, making explicit a meaning that had been too dependent on English-language recovery vocabulary. “Shifting the balance of power” became “Moving toward an equal partnership”, replacing an abstract and potentially adversarial formulation with wording that was more relational and consistent with shared decision-making language. “Enabling different relationships” became “Connecting and working together”, removing ambiguity around what “different” meant. “Facilitating personal growth” became “Facilitating recovery”, avoiding a potentially culturally loaded assumption that improvement must take the form of upward growth. Similar refinements occurred within the Fidelity Measure. “Professionals” became “staff”, reducing dependence on systems in which “professional” implies formal registration or status. “Tailored to the student” was reframed to emphasise flexibility, reasonable adjustments and feasibility within local context and resources, rather than assuming a uniformly well-resourced service infrastructure. “Commitment to recovery” was expanded to specify that recovery refers to personal and social recovery, not only clinical recovery or symptom reduction. Collectively, these changes moved the framework away from locally saturated shorthand and toward wording that was more explicit, relational and transferable. These findings have implications beyond RCs. Mental health tools developed in high-income Anglophone contexts may be presented as broadly applicable while still embedding culturally and institutionally specific assumptions about service organisation, professional identity, rights discourse and interpersonal style. This is a general risk when any national or regional model becomes influential internationally. The present study shows why explicit de-centring is needed before such tools are adopted globally. For cross-cultural psychiatry, this supports a move away from assuming that transportability follows automatically from translation, and towards testing conceptual equivalence, semantic clarity and implementation feasibility across settings. 2 It also speaks to current debates on racialised and culturally unequal mental healthcare, where seemingly neutral service language can obscure the standpoint from which it was produced. 24 De-centring therefore has both methodological and ethical value: it improves interpretability while reducing the risk that culture-specific assumptions are universalised. The revised wording provides the Global RCM v1 and Global RFM v1 for research, service development and dialogue across international RC settings. These v1 tools are not intended to replace the original England-developed versions, but to retain interoperability with them while supporting more consistent interpretation across countries. This matters because international comparison requires enough continuity to compare findings, but enough de-centring to avoid imposing source-context assumptions uncritically. The Global RCM v1 and Global RFM v1 may therefore support cross-national RC research, local adaptation work, and policy dialogue in countries seeking to build recovery-oriented services without simply importing Anglophone terminology. In this sense, the revised tools function as resources for international learning rather than one-way model transfer. This study has several strengths. Its originality lies in applying an explicit cross-cultural de-centring approach to internationally used RC tools that had previously remained anchored in an English conceptual frame. Many cross-cultural mental health studies adopt frontal comparison, 25 in which tools or interventions developed in one source context are assessed against one or more target contexts. This can be useful for identifying adaptation needs, but may also reproduce a source-versus-target logic in which non-Anglophone settings are mainly understood by how they differ from an Anglophone reference point. By contrast, this study enabled a more inclusive lateral comparison across multiple RC settings, examining how concepts were interpreted from multiple standpoints and generating findings with broader international relevance. Its rigour lies in the mixed-method design: an international Delphi across 20 countries, predefined quantitative consensus thresholds, corpus-informed linguistic analysis of free-text responses, and iterative research team consultation to ensure that revisions were both evidence-informed and conceptually coherent. Its impact potential lies in producing the Global RCM v1 and Global RFM v1, which can be used more credibly in international RC research and service development while retaining interoperability with the England-developed RCM and RFM, allowing comparison between studies using either version. The study also has limitations. The sample was international but necessarily limited to countries where RCs exist or where relevant expertise was available; therefore, the study should not be interpreted as representing all world regions. However, given that RCs have been identified in 31 countries, participation from 20 countries represents substantial coverage of the current international RC community. Participation was still uneven across regions and countries, and some RC settings were represented by more panellists than others. The consultation was conducted in English, which may have favoured panellists more comfortable engaging with English-language recovery discourse, although this also made visible some of the linguistic assumptions that the study sought to examine. Round 2 established consensus on understandability and cultural appropriateness of the revised wording, but did not test translated versions or implementation in each context. Finally, because panellists were people already engaged with RCs, the study does not capture views from all stakeholders who may encounter these tools in wider policy or service environments. Future work should test how the Global RCM v1 and Global RFM v1 perform in translated versions and specific national settings, including whether further adaptation is needed at the level of examples, scoring or guidance for interpretation. Comparative work could also examine whether some items continue to function differently across linguistic regions, service systems or legal-political contexts. More broadly, the study supports a research agenda in cross-cultural psychiatry that treats tool adaptation as a substantive epistemic task rather than a final technical step. In conclusion, this Delphi study suggests that Recovery College principles are broadly valued internationally, but that wording developed in a high-income Anglophone context required de-centring before it could become more globally usable. The Global RCM v1 and Global RFM v1 therefore represent not a departure from the Recovery College model, but a clearer and more internationally applicable articulation of it. Methods This study followed the ACcurate COnsensus Reporting Document (ACCORD) reporting guideline for consensus methods (Supplementary File 2). 26 A study protocol was published in advance, 27 and the study materials and anonymised underlying data are openly available on the Open Science Framework (OSF; http://doi.org/10.17605/OSF.IO/CUAES). Ethical approval was obtained from the King’s College London Research Ethics Psychiatry, Nursing and Midwifery Subcommittee (reference: MRM-24/25-47085). Study design We conducted a two-round international Delphi study to examine the cross-cultural applicability of the RCM and RFM. The study followed Belton’s six-step Delphi methodology 28 and used a decentring approach to cross-cultural research, 29 which seeks to review and refine tools developed in one cultural context to improve their relevance across others. The Delphi aimed to identify: (1) the perceived importance of each RCM mechanism and RFM non-modifiable item, (2) the perceived cultural difficulty of achieving them, (3) recommended wording changes to improve cross-cultural applicability, and (4) which response category for each RFM modifiable item best aligned with local cultural context. Participants and recruitment Participants were recruited through the RECOLLECT International Research Consortium, 30 which, at the time of the study, spanned 31 countries across Africa, Asia, Europe, North America, South America and Oceania. Country leads were invited to participate as panellists and to recruit additional eligible panellists within their national networks. Eligibility criteria were: age 18 years or older and experience of a RC in any role. To avoid over-representation from individual countries, recruitment was monitored by country, and country leads were asked to stop recruitment once five responses had been received, although any additional responses already submitted were retained. Consent was implied through survey completion, in line with standard Delphi practice. 31 Panellist anonymity was maintained through an individual online survey design, without access to the identities or responses of other panellists. This reduced the risk that panellists, particularly those with less experience or from countries where RCs are still emerging, might feel inhibited in expressing their views. 28 Names were given optionally for acknowledgement purposes and were accessible only to the Delphi administrator (YK), who had no voting rights in the consensus process. Survey development and procedure The draft survey was created by YK, reviewed by the research team, and piloted by country leads from Brazil, Estonia, Germany, Japan and Thailand to improve clarity, linguistic accessibility, suitability for online completion, and cross-cultural relevance. Revisions following piloting included adding clarifying examples to potentially ambiguous terms, and incorporating optional free-text comment boxes alongside quantitative responses. The finalised survey was used in Round 1 in English. Panellists rated the four RCM mechanisms and seven RFM non-modifiable items for (a) importance and (b) cultural difficulty of achieving them in their context. Rating used five-point Likert scales (1=”Not at all” to 5=”Very important/difficult”), and for each item panellists could also suggest wording changes to improve cultural applicability. Panellists then indicated for each of the five RFM modifiable items which of the two response categories better aligned with the cultural context of RCs in their country, and could provide wording suggestions. Round 1 remained open from July to November 2025 (16 weeks), closing after seven consecutive days without a new response. The Round 1 survey was then modified as described in the next section, to create the Round 2 survey. Round 2 remained open from January to February 2026 (5 weeks). Fortnightly reminders were sent to support retention. No financial incentives were used, as these may undermine the quality of panellist input in Delphi studies. 28 Panellists were offered social recognition through acknowledgement in the final paper. 32 Analysis For quantitative importance and cultural difficulty ratings, consensus was predefined as 70% or more of responses falling within the same category after collapsing the original five-point scale into three categories: responses of 1 and 2 were combined, 3 was retained as a middle category, and 4 and 5 were combined. This prespecified approach was used to reduce potential cross-cultural response bias, particularly variation in the use of extreme response options across cultural groups, and was supported by pilot findings indicating that a three-point structure would preserve meaningful distinctions while improving comparability across cultural groups. 27 Descriptive statistics were used to summarise responses, including frequencies, proportions, and medians (interquartile ranges (IQRs)). For RFM modifiable items, frequencies and proportions were calculated for Type 1 versus Type 2 responses. Items not reaching consensus in Round 1, or generating substantial qualitative concern, were revised and re-presented in Round 2. Free-text responses in Round 1 were analysed using corpus-informed linguistic analysis, a software-assisted approach for examining large collections of digitised text, to identify recurrent terms and themes, patterns of language use, and ambiguities that informed wording changes to improve cross-cultural applicability. This approach differed from the published protocol, 27 which had proposed descriptive content analysis, because the responses consisted of brief, item-specific comments from panellists across 20 countries, making a systematic comparison of wording patterns especially appropriate. Open-text responses were grouped into datasets for each question type: cultural difficulty, components and mechanisms; importance, components and mechanisms; and cultural fit and wording. Keywords, defined as words statistically salient in one dataset compared with a reference dataset, were retrieved for each response dataset using the enTenTen21 corpus as reference (~52 billion words, available in Sketch Engine, the software used for the analysis). Salient themes and insights into language use and preferences were identified inductively by examining concordance lines, that is, strings of text displaying terms of interest in context. Findings were reviewed by the research team, including the developers of the RCM and RFM, and were used to generate revised wording for Round 2. In Round 2, panellists were asked whether the proposed revisions were appropriate in their contexts and, if not, to suggest further changes. Suggested revisions were reviewed and integrated by YK in discussion with co-authors. All panellists were sent the finalised Global RCM v1 and Global RFM v1 after Round 2. Declarations Authors’ contributions Conceptualisation: YK; Methodology: YK, SVL, VL, CH, MS; Validation: YK, SVL, DN, RH, PL, JO, MCRA, YM, UB, JPP, VR, MN, TJ, VL, ST, AR, VK, AK, PB, CH, MS; Investigation: YK, SVL; Writing – original draft: YK, SVL; Writing – review & editing: YK, SVL, DN, RH, PL, JO, MCRA, YM, UB, JPP, VR, MN, TJ, VL, ST, AR, VK, AK, PB, CH, MS; Funding acquisition: CH, MS. Acknowledgements We thank Marijn Antens, Simone Arbour, Ana Aznar Martínez, Anders Thorkil Bechgaard, Anne Mette Billekop, Lily Chan, Danielle Coppleson, Sally Easton, Matthew Jonathan Ellis, Daniela Fischer, Vinuri Fernando, Andrew Foster, Kumpon Kawkome, Kenji Kameda, Megumi Kurokami, Aya Kuroda, Ossi Kukkamäki, Ben Hart, Christine Larsen, Klaudia Lepska, Pernilla Lemvall, Living Recovery Center (Thailand), Sue Macdonald, Chiharu Nakayama, Andrew Nicholls, Jared Omundo, Apirak Paenmunin, Jörn Petersen, Supatchakarn Piros, Geoff Turner, Flavio Trolese, Paola Santo, Gianpaolo Scarsato, Erica Wan, Louisa Whorrod, Alex Thomas, Aurelie Tinland, Jonna Tolonen, Wilasinee Pinyowit, Kruawon Tiengtom, Asuka Yoshimi and Yoshie Morita for their support with this study. MS acknowledges the support of NIHR Nottingham Biomedical Research Centre. Funding This study is funded by the National Institute for Health and Care Research (Recovery Colleges Characterisation and Testing (RECOLLECT) 2, ID: NIHR200605). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests The authors declare no competing interests. Patient consent for publication Not applicable. Ethics approval Ethical approval was obtained from the King’s College London Research Ethics Psychiatry, Nursing and Midwifery Subcommittee (reference: MRM-24/25-47085). Data availability statement Study materials and underlying data are openly available on the Open Science Framework: http://doi.org/10.17605/OSF.IO/CUAES. References Whitley, R., Shepherd, G. & Slade, M. Recovery colleges as a mental health innovation. World psychiatry : official journal of the World Psychiatric Association (WPA) 18 , 141-142, doi:https://dx.doi.org/10.1002/wps.20620 (2019). Henderson, C. et al. Social inclusion of people with severe mental illness: a review of current practices, evidence and unmet needs, and future directions. World Psychiatry 25 , 56-82, doi:https://doi.org/10.1002/wps.70031 (2026). Hayes, D. et al. Organisational and student characteristics, fidelity, funding models, and unit costs of recovery colleges in 28 countries: a cross-sectional survey. The lancet. Psychiatry 10 , 768-779, doi:https://dx.doi.org/10.1016/S2215-0366(23)00229-8 (2023). Lin, E. et al. Developing an evaluation framework for assessing the impact of recovery colleges: protocol for a participatory stakeholder engagement process and cocreated scoping review. BMJ open 12 , e055289, doi:https://dx.doi.org/10.1136/bmjopen-2021-055289 (2022). Theriault, J., Lord, M. M., Briand, C., Piat, M. & Meddings, S. Recovery colleges after a decade of research: a literature review. Psychiatric Services 71 , 928-940, doi:10.1176/appi.ps.201900352 (2020). Crowther, A. et al. The impact of Recovery Colleges on mental health staff, services and society. Epidemiology and psychiatric sciences 28 , 481-488, doi:https://dx.doi.org/10.1017/S204579601800063X (2019). Ronaldson, A. et al. The impact of Recovery College enrolment on health service use and patient outcomes: retrospective matched cohort study using routinely collected data. medRxiv , 2025.2012.2009.25341905, doi:10.64898/2025.12.09.25341905 (2025). Deloitte. The shift to prevention: Realising the socio-economic potential , (2025). Hayes, D. et al. Recovery Colleges Characterisation and Testing in England (RECOLLECT): rationale and protocol. BMC Psychiatry 22 , doi:10.1186/s12888-022-04253-y (2022). Toney, R. et al. Development and Evaluation of a Recovery College Fidelity Measure. Canadian journal of psychiatry. Revue canadienne de psychiatrie 64 , 405-414, doi:https://dx.doi.org/10.1177/0706743718815893 (2019). Hayes, D. et al. Evidence-based Recovery Colleges: developing a typology based on organisational characteristics, fidelity and funding. Social Psychiatry and Psychiatric Epidemiology 59 , 759-768, doi:10.1007/s00127-023-02452-w (2024). Kotera, Y. et al. Cross-Cultural Insights from Two Global Mental Health Studies: Self-Enhancement and Ingroup Biases. International Journal of Mental Health and Addiction 23 , 3572-3581, doi:10.1007/s11469-024-01307-y (2025). Griner, D. & Smith, T. B. Culturally adapted mental health intervention: A meta-analytic review. Psychotherapy (Chic) 43 , 531-548, doi:10.1037/0033-3204.43.4.531 (2006). Benish, S. G., Quintana, S. & Wampold, B. E. Culturally adapted psychotherapy and the legitimacy of myth: a direct-comparison meta-analysis. J Couns Psychol 58 , 279-289, doi:10.1037/a0023626 (2011). Sjöberg, A., Liv, P. & Lindström, M. Effect of Everyday Life Rehabilitation on recovering quality of life in individuals with serious mental illness in supported accommodation: a pragmatic cluster randomised controlled trial. BMJ Mental Health 28 , e301757, doi:10.1136/bmjment-2025-301757 (2025). Kuek, J. H. L., Raeburn, T. & Wand, T. Asian perspectives on personal recovery in mental health: a scoping review. J Ment Health 32 , 517-533, doi:10.1080/09638237.2020.1818709 (2023). Kotera, Y. et al. Cross-cultural Comparison of Recovery College Implementation Between Japan and England: Corpus-based Discourse Analysis. International Journal of Mental Health and Addiction 23 , 4362-4386, doi:10.1007/s11469-024-01356-3 (2025). Kotera, Y. et al. Recovery College Advertising in the Netherlands: A Dutch-English-Japanese Corpus-Based Comparison and Its Implications for Quality-of-Life Orientation. [preprint] , doi:10.1101/2025.08.13.25333552 (2025). Kotera, Y. et al. 28-country global study on associations between cultural characteristics and Recovery College fidelity. npj Mental Health Research 3 , doi:10.1038/s44184-024-00092-9 (2024). Kotera, Y. et al. Cultural influences on fidelity components in recovery colleges: a study across 28 countries and territories. General Psychiatry 38 , e102010, doi:https://dx.doi.org/10.1136/gpsych-2024-102010 (2025). Hofstede, G., Hofstede, G. J. & Minkov, M. Cultures and organizations: Software of the mind . 3rd edn, (McGraw-Hill Education, 2010). Nazarian, A., Ranjbaran, A., Afifi, M., Atkinson, P. & Sadeghilar, N. Cross-cultural examination of effects of ethical leadership of independent hotels based on Social Exchange Theory. International Journal of Cross Cultural Management 24 , 463-484, doi:10.1177/14705958241270817 (2024). Textbook of Cultural Psychiatry . 2 edn, (Cambridge University Press, 2018). Stepanova, E. et al. "I am not a priority": ethnic minority experiences of navigating mental health support and the need for culturally sensitive services during and beyond the pandemic. BMJ Mental Health 28 , e301481, doi:10.1136/bmjment-2024-301481 (2025). Schnegg, M. & Lowe, E. D. Comparing Cultures: Innovations in Comparative Ethnography . (Cambridge University Press, 2020). Gattrell, W. T. et al. ACCORD (ACcurate COnsensus Reporting Document): A reporting guideline for consensus methods in biomedicine developed via a modified Delphi. PLoS Med 21 , e1004326, doi:10.1371/journal.pmed.1004326 (2024). Kotera, Y. et al. Enhancing cross-cultural applicability in recovery colleges: A global Delphi study protocol. PloS one 20 , e0332729, doi:https://dx.doi.org/10.1371/journal.pone.0332729 (2025). Belton, I., MacDonald, A., Wright, G. & Hamlin, I. Improving the practical application of the Delphi method in group-based judgment: A six-step prescription for a well-founded and defensible process. Technological Forecasting and Social Change 147 , 72-82, doi:https://doi.org/10.1016/j.techfore.2019.07.002 (2019). van de Vijver, F. J. R. & Leung, K. Methods and Data Analysis for Cross-Cultural Research . 2 edn, (Cambridge University Press, 2021). Recovery Research Team. RECOLLECT International Research Consortium (RIRC) , (2025). Toma, C. L. & Picioreanu, I. The Delphi Technique: Methodological Considerations and the Need for Reporting Guidelines in Medical Journals. International journal of public health research 4 , 47 (2016). Khodyakov, D., Grant, S., Kroger, J. & Bauman, M. (RAND Corporation, Santa Monica, CA, USA, 2023). Additional Declarations There is NO Competing Interest. Supplementary Files SF16RFMACCORDLingRevR1to2PostR2Final.docx Supplementary files 1 to 6 Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9614758","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":634566673,"identity":"34c9169a-45ee-4ecc-883a-86d038f629cc","order_by":0,"name":"Yasuhiro Kotera","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0002-0251-0085","institution":"University of Nottingham","correspondingAuthor":true,"prefix":"","firstName":"Yasuhiro","middleName":"","lastName":"Kotera","suffix":""},{"id":634566674,"identity":"306c05d3-01b2-4628-9413-51454a2aaa60","order_by":1,"name":"Sara Vilar-Lluch","email":"","orcid":"","institution":"Cardiff University","correspondingAuthor":false,"prefix":"","firstName":"Sara","middleName":"","lastName":"Vilar-Lluch","suffix":""},{"id":634566675,"identity":"36742f88-c2f2-4bb7-b8fd-36ec6922e341","order_by":2,"name":"Dagmar Narusson","email":"","orcid":"","institution":"University of Tartu","correspondingAuthor":false,"prefix":"","firstName":"Dagmar","middleName":"","lastName":"Narusson","suffix":""},{"id":634566676,"identity":"7668508d-f7f1-4fd5-97ce-fd6fd99b4dc5","order_by":3,"name":"Ramona Hiltensperger","email":"","orcid":"https://orcid.org/0000-0003-2544-4188","institution":"Ulm University","correspondingAuthor":false,"prefix":"","firstName":"Ramona","middleName":"","lastName":"Hiltensperger","suffix":""},{"id":634566677,"identity":"055fa53f-b7b5-4bf3-a8eb-63cee8f598e8","order_by":4,"name":"Peanchanan Leah","email":"","orcid":"","institution":"Mahidol University","correspondingAuthor":false,"prefix":"","firstName":"Peanchanan","middleName":"","lastName":"Leah","suffix":""},{"id":634566678,"identity":"879fdc2c-1878-4ecc-b7a7-197564256454","order_by":5,"name":"Jose Orsi","email":"","orcid":"","institution":"Universidade Federal de São Paulo","correspondingAuthor":false,"prefix":"","firstName":"Jose","middleName":"","lastName":"Orsi","suffix":""},{"id":634566679,"identity":"074aa0f5-2d3e-461f-a09b-d56140017fbf","order_by":6,"name":"Mário Andrade","email":"","orcid":"","institution":"Universidade Federal de São Paulo","correspondingAuthor":false,"prefix":"","firstName":"Mário","middleName":"","lastName":"Andrade","suffix":""},{"id":634566680,"identity":"7bd4bfc1-6df5-4b2f-86e3-20a453eb0c30","order_by":7,"name":"Yuki Miyamoto","email":"","orcid":"","institution":"The University of Tokyo","correspondingAuthor":false,"prefix":"","firstName":"Yuki","middleName":"","lastName":"Miyamoto","suffix":""},{"id":634566681,"identity":"3edb62df-acb8-42eb-a110-61a8b39f67f2","order_by":8,"name":"Ulrika Bejerholm","email":"","orcid":"","institution":"Lund University","correspondingAuthor":false,"prefix":"","firstName":"Ulrika","middleName":"","lastName":"Bejerholm","suffix":""},{"id":634566682,"identity":"16fee828-aa3c-4809-88f3-c6ffd40caa42","order_by":9,"name":"Jana Pluhaříková Pomajzlová","email":"","orcid":"","institution":"Centre for Mental Health Care Development","correspondingAuthor":false,"prefix":"","firstName":"Jana","middleName":"Pluhaříková","lastName":"Pomajzlová","suffix":""},{"id":634566683,"identity":"e7e5737d-b383-42e6-a33f-aa7fe246ea2a","order_by":10,"name":"Vlasta Řezníková","email":"","orcid":"","institution":"College of Polytechnics Jihlava","correspondingAuthor":false,"prefix":"","firstName":"Vlasta","middleName":"","lastName":"Řezníková","suffix":""},{"id":634566684,"identity":"a9f8f6d5-0265-439f-a293-27087d501e72","order_by":11,"name":"Mariam Nabasaba","email":"","orcid":"","institution":"King's College London","correspondingAuthor":false,"prefix":"","firstName":"Mariam","middleName":"","lastName":"Nabasaba","suffix":""},{"id":634566685,"identity":"8025e223-ed08-499d-b911-1fe09cad5705","order_by":12,"name":"Tesnime Jebara","email":"","orcid":"","institution":"King's College London","correspondingAuthor":false,"prefix":"","firstName":"Tesnime","middleName":"","lastName":"Jebara","suffix":""},{"id":634566686,"identity":"f9f2d165-f287-40a9-978a-0d4b1fb53f61","order_by":13,"name":"Vanessa Lawrence","email":"","orcid":"","institution":"King's College London","correspondingAuthor":false,"prefix":"","firstName":"Vanessa","middleName":"","lastName":"Lawrence","suffix":""},{"id":634566687,"identity":"3d65290d-1bd3-4fb4-9ea4-39e925098d9c","order_by":14,"name":"Simran Takhi","email":"","orcid":"","institution":"University of Nottingham","correspondingAuthor":false,"prefix":"","firstName":"Simran","middleName":"","lastName":"Takhi","suffix":""},{"id":634566688,"identity":"6de5f5db-1889-46bf-a101-0557a4bdf947","order_by":15,"name":"Amy Ronaldson","email":"","orcid":"","institution":"King's College London","correspondingAuthor":false,"prefix":"","firstName":"Amy","middleName":"","lastName":"Ronaldson","suffix":""},{"id":634566689,"identity":"ef603d89-a12e-47ae-847a-e1ddfa985370","order_by":16,"name":"Vanessa Kellermann","email":"","orcid":"","institution":"King's College London","correspondingAuthor":false,"prefix":"","firstName":"Vanessa","middleName":"","lastName":"Kellermann","suffix":""},{"id":634566690,"identity":"e0e6f5d3-7424-4c74-b6cd-1a375c541423","order_by":17,"name":"Agnieszka Kapka","email":"","orcid":"","institution":"King's College London","correspondingAuthor":false,"prefix":"","firstName":"Agnieszka","middleName":"","lastName":"Kapka","suffix":""},{"id":634566691,"identity":"999d2bac-a856-4c36-95b3-973e9503a134","order_by":18,"name":"Peter Bates","email":"","orcid":"","institution":"RECOLLECT Lived Experience Advisory Panel","correspondingAuthor":false,"prefix":"","firstName":"Peter","middleName":"","lastName":"Bates","suffix":""},{"id":634566692,"identity":"b3367140-4dff-44ee-b618-020d37b0bfe7","order_by":19,"name":"Claire Henderson","email":"","orcid":"","institution":"King's College, Institute of Psychiatry Psychology \u0026 Neuroscience","correspondingAuthor":false,"prefix":"","firstName":"Claire","middleName":"","lastName":"Henderson","suffix":""},{"id":634566693,"identity":"8515ea09-8d46-4664-8c69-e72b4d611e01","order_by":20,"name":"Mike Slade","email":"","orcid":"https://orcid.org/0000-0001-7020-3434","institution":"University of Nottingham","correspondingAuthor":false,"prefix":"","firstName":"Mike","middleName":"","lastName":"Slade","suffix":""}],"badges":[],"createdAt":"2026-05-05 06:50:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9614758/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9614758/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108805350,"identity":"6dc7d58d-7091-4e74-9bca-5fd8211cae40","added_by":"auto","created_at":"2026-05-08 15:25:40","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":526425,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9614758/v1/5ae2da97-a48a-46d2-b65f-2841d776d08d.pdf"},{"id":108567659,"identity":"94b666d0-e0b3-4e06-a404-2cf0ffb1ca1d","added_by":"auto","created_at":"2026-05-06 05:20:50","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":63491,"visible":true,"origin":"","legend":"Supplementary files 1 to 6","description":"","filename":"SF16RFMACCORDLingRevR1to2PostR2Final.docx","url":"https://assets-eu.researchsquare.com/files/rs-9614758/v1/d1a6ca2d3391b38df33bd20d.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Cross-cultural de-centring of Recovery College tools: a 20-country Delphi study","fulltext":[{"header":"Introduction","content":"\u003cp\u003eRecovery Colleges (RCs) are an innovative approach\u0026nbsp;to mental health support that combines co-production and adult learning to promote personal recovery and social inclusion.\u003csup\u003e1,2\u003c/sup\u003e Established in England in 2009, RCs have since expanded internationally. A previous global review identified 221 RCs in 28 countries in 2022,\u003csup\u003e3\u003c/sup\u003e and we are aware of further growth to 264 RCs worldwide by 2026. RCs provide inclusive educational environments in which people with mental health problems, carers, staff, and people in the wider community (referred to as “students”) learn together, develop skills and build support networks. Through self-directed learning and shared expertise, RCs support fuller social participation, greater self-management and personal recovery.\u003csup\u003e4\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eEmerging evidence indicates that RCs can improve student outcomes, including self-esteem, quality of life and internalised stigma,\u003csup\u003e5\u003c/sup\u003e and benefit staff through increased motivation and skill development.\u003csup\u003e6\u003c/sup\u003e A large retrospective matched cohort study involving 1,435 students and 4,665 controls found reduced subsequent mental health service use and a £5,028 greater reduction in total costs per student at 12 months, equivalent to an estimated 8.4:1 return on investment.\u003csup\u003e7\u003c/sup\u003e Uncontrolled before-and-after analyses in the same evaluation also showed improved symptoms and functioning.\u003csup\u003e7\u003c/sup\u003e These economic findings compare favourably with a 2025 estimate of an 8:1 socio-economic return for preventive healthcare interventions.\u003csup\u003e8\u003c/sup\u003e However, operational guidance for RCs has only recently become more standardised through the Recovery Colleges Characterisation and Testing (RECOLLECT) programme (NIHR200605).\u003c/p\u003e\n\u003cp\u003eTwo widely-used outputs from RECOLLECT are the RECOLLECT Change Model (RCM) and the RECOLLECT Fidelity Measure (RFM). The RCM identifies four mechanisms through which RCs are proposed to work: empowering environment; shifting the balance of power; enabling different relationships; and facilitating personal growth.\u003csup\u003e9\u003c/sup\u003e The RFM is a 12-item tool assessing fidelity to core RC principles and best practice.\u003csup\u003e10\u003c/sup\u003e The 12 items comprise seven non-modifiable items, representing core principles expected to be present across Recovery Colleges, and five modifiable items, representing features that may appropriately vary by local context. Non-modifiable items include equality, adult learning, tailoring to the student, co-production, social connectedness, community focus and commitment to recovery. Modifiable items concern accessibility, location, distinctiveness of course content, strength-based orientation and progressiveness; for example, whether access is universal or targeted, whether the setting is Recovery College-specific or community-integrated, and whether course content is distinct from or integrated with clinical services (Supplementary File 1). Developed in England through an empirical programme of work, these tools drew on a literature review, consultation with 77 international experts, interviews with 10 Recovery College managers, refinement with 44 students, trainers and managers across 3 sites, and evaluation involving 60 professional stakeholders.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe international growth of RCs raises an important question: whether tools developed in one cultural context can be applied appropriately across others. This concern is particularly relevant because most RC research has been conducted in England, where approximately one third of all RCs are based,\u003csup\u003e11\u003c/sup\u003e and in other high-income Global North settings, limiting understanding of how RC principles and language operate across more diverse cultural, linguistic and service systems.\u003csup\u003e3,12\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eCultural fit matters. A meta-analysis of 76 studies involving 25,225 participants found that culturally adapted mental health interventions showed a moderately strong advantage over unadapted approaches (d=.45).\u003csup\u003e13\u003c/sup\u003e A direct-comparison meta-analysis also found that culturally adapted psychotherapy outperformed unadapted psychotherapy (d=0.32).\u003csup\u003e14\u003c/sup\u003e Together, these findings indicate that cultural adaptation is not a peripheral concern: the language, examples and assumptions embedded in mental health interventions and implementation tools may materially affect whether they are understood, accepted and effective. Because the RCM and RFM are intended to be used across diverse RC contexts, the key methodological challenge was not only translation or local adaptation, but cross-cultural interoperability: preserving shared conceptual meaning while allowing wording and examples to fit different cultural and service contexts. A consensus exercise was therefore chosen to integrate international stakeholder judgement on how the RCM mechanisms and RFM items were interpreted across contexts, where agreement or disagreement indicated culturally different understandings, and how the wording should be refined to improve cross-cultural interoperability.\u003c/p\u003e\n\u003cp\u003eThis study aimed to identify the level of cultural influence on the RCM mechanisms and RFM items, and to develop globally interoperable versions of these tools.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eA total of 54 panellists from 20 countries completed Round 1. Participants included 22 RC managers (40.7%), 10 RC researchers (18.5%), 9 RC peer trainers (16.7%), 7 RC non-peer trainers (13.0%), and 6 other RC staff (11.1%). In terms of RC experience, 13 panellists (24.1%) had 7+ years\u0026rsquo; experience, 13 (24.1%) had 4\u0026ndash;6 years, 22 (40.7%) had 1\u0026ndash;3 years, and 6 (11.1%) had less than 1 year. Geographically, panellists were mainly from Western Europe (42.6%) and Asia (29.6%), with additional representation from Central Europe (11.1%), Oceania (5.6%), North America (5.6%), South America (3.7%), and Eastern Europe (1.9%). In Round 2, 41 panellists (75.9%) from 15 countries participated. Geographical representation remained strongest from Western Europe (46.3%) and Asia (26.8%), followed by Central Europe (9.8%), North America (7.3%), Oceania (4.9%), and South America (4.9%). Table 1 presents the panellist characteristics.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eTable 1. Panellist characteristics (n=54)\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eCountry\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003ePrimary role\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eYears of experience\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eR2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAustralia\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC staff\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7+ years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAustralia\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBrazil\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBrazil\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC researcher\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eCanada\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC researcher\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7+ years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eCanada\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eCanada\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4\u0026ndash;6 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eCzech Republic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4\u0026ndash;6 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eCzech Republic / Slovakia\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC researcher\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4\u0026ndash;6 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eDenmark\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7+ years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eDenmark\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC non-peer trainer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7+ years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eDenmark\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC peer trainer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eEngland\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC researcher\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eEngland\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7+ years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eEngland\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eEngland\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7+ years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eEngland\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4\u0026ndash;6 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eEngland\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC staff\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4\u0026ndash;6 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eEngland\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7+ years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eEngland\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC peer trainer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLess than 1 year\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eEngland\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC peer trainer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4\u0026ndash;6 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eEstonia\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC researcher\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLess than 1 year\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eFinland\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC researcher\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eFinland\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4\u0026ndash;6 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eFinland\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC non-peer trainer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLess than 1 year\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eFrance\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC researcher\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7+ years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eGermany\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC researcher\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eGermany\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7+ years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eGermany\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC non-peer trainer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eHong Kong\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4\u0026ndash;6 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eHong Kong\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC researcher\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7+ years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eItaly\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eItaly\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7+ years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eJapan\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC non-peer trainer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7+ years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eJapan\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC non-peer trainer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4\u0026ndash;6 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eJapan\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC peer trainer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eJapan\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC researcher\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eJapan\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4\u0026ndash;6 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eJapan\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eJapan\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC non-peer trainer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eJapan\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC non-peer trainer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4\u0026ndash;6 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNetherlands\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC researcher\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLess than 1 year\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNew Zealand\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC staff\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSpain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7+ years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSweden\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC researcher\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSweden\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLess than 1 year\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSweden\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSwitzerland\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC manager\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eThailand\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC peer trainer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eThailand\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC peer trainer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7+ years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eThailand\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC peer trainer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLess than 1 year\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eThailand\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC peer trainer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4\u0026ndash;6 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eThailand\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC peer trainer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4\u0026ndash;6 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eThailand\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRC staff\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u0026ndash;3 years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e✓\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eR2 = participated in the Round 2 survey. RC = Recovery College. RC non-peer trainer includes roles such as psychologist, nurse, social worker, or educator.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRound 1 quantitative findings\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll four RCM mechanisms reached consensus for importance, with 92.6%\u0026ndash;98.1% of panellists rating them in the 4\u0026ndash;5 category after collapsing the five-point scale into three categories (Table 2). Six of the seven RFM non-modifiable items also reached consensus for importance (70.4%\u0026ndash;100.0%). The exception was tailored to the student, which fell just below the threshold (37/54, 68.5%). Median importance ratings were uniformly high (median 4\u0026ndash;5).\u003c/p\u003e\n\u003cp\u003eBy contrast, consensus on cultural difficulty was more limited. Crucially, no RCM mechanism reached consensus for low difficulty (1\u0026ndash;2 category). Among the RFM non-modifiable items, only Learning, Social connectedness, and Commitment to recovery reached consensus for low difficulty, with 70.4%, 70.4%, and 75.9% of panellists respectively rating them in the 1\u0026ndash;2 category. Median difficulty ratings ranged from 2 to 3 across items, indicating that achieving these standards involves significant cultural navigation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eTable 2. Round 1 ratings for importance and cultural difficulty of RCM mechanisms and RFM non-modifiable items (N=54)\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eItem\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eImportance\u0026nbsp;\u003c/strong\u003e4\u0026ndash;5 n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eImportance\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eMedian (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eImportance consensus met\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eDifficulty\u0026nbsp;\u003c/strong\u003e1\u0026ndash;2 n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eDifficulty\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eMedian (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eDifficulty consensus met\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eRCM mechanisms\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eEmpowering environment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e51 (94.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e33 (61.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e2 (2\u0026ndash;3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eShifting the balance of power\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e50 (92.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e22 (40.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e3 (2\u0026ndash;4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eEnabling different relationships\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e52 (96.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e28 (51.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e2 (2\u0026ndash;3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eFacilitating personal growth\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e53 (98.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (5\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e34 (63.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e2 (2\u0026ndash;3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eRFM non-modifiable items\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e1. Valuing equality\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e54 (100.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (5\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e33 (61.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e2 (2\u0026ndash;3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e2. Learning\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e48 (88.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e38 (70.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e2 (2\u0026ndash;3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e3. Tailored to the student\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e37 (68.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e4 (3\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e20 (37.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e3 (2\u0026ndash;3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e4. Co-production of the RC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e49 (90.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e27 (50.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e2 (2\u0026ndash;3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e5. Social connectedness\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e50 (92.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e38 (70.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e2 (2\u0026ndash;3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e6. Community focus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e38 (70.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e4 (3\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e24 (44.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e3 (2\u0026ndash;4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e7. Commitment to recovery\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e51 (94.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (5\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e41 (75.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e2 (1\u0026ndash;2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eConsensus was predefined as \u0026ge;70% of responses falling within the same collapsed category. For importance, consensus was assessed in the \u003cstrong\u003e4\u0026ndash;5\u003c/strong\u003e category. For difficulty, consensus was assessed in the \u003cstrong\u003e1\u0026ndash;2\u003c/strong\u003e category. Medians and IQRs (Interquartile Ranges) are based on the original five-point scale.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRound 1 qualitative findings informing revision\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo understand how panellists interpreted the items across settings and to identify wording that could improve cross-cultural applicability, free-text responses from Round 1 (totalling 9,541 words) were analysed using corpus-informed linguistic analysis. This analysis extracted \u0026lsquo;keywords\u0026rsquo; (terms significantly salient in the focus dataset in comparison to a larger one of general English language) for each dataset of the free-text responses to identify recurrent terms and patterns of language use, and inductively identify themes. Examination of the keywords in context of use can identify ambiguities and points of tension in how panellists understood and applied the RC wording/terminology. Table 3 presents the main cross-cultural issues identified (Supplementary File 3 for more details).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eTable 3. Main cross-cultural issues identified in Round 1 free-text responses and corresponding wording revision\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\" class=\"fr-table-selection-hover\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eItem\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003ePanellist feedback\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eRevision made\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eEmpowering environment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eLacked a clear equivalent in some contexts; meaning was inconsistent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eEmphasised strengths, autonomy, choice and control\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eShifting the balance of power\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eFelt abstract, confrontational or hard to realise; cultural and structural barriers noted\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eReframed as equal partnership\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eEnabling different relationships\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eUnclear what relationships were meant and with whom\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRevised to focus on connecting and working together\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eFacilitating personal growth\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eExpectation of upward progress did not fit recovery in some contexts\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRecentred on recovery rather than growth\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eValuing equality\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eTension between equality and individual needs; \u0026ldquo;student\u0026rdquo; label queried\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eEmphasised accessibility, equity and clearer role descriptors\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eLearning\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eAdult learning approaches were unfamiliar in some settings; terminology was too jargonistic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eSimplified and clarified in non-jargonistic terms\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eTailored to the student\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eInterpreted variously as person-centred flexibility or cultural adaptation; resource constraints noted\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eClarified as adapting to student needs within local context\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eCo-production of the RC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eMeaning unclear; panellists requested explanation or examples\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eDescribed shared decision-making and contributor roles more explicitly\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eSocial connectedness\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eAmbiguity between internal RC relationships and wider community links\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eDistinguished connectedness within the RC and beyond\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eCommunity focus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eUnclear what kind of community was meant; organisational barriers noted\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eMade intended meaning of community more explicit\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eCommitment to recovery\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eOften understood as clinical rather than personal or social recovery\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eClarified intended meaning of recovery\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eFree-text feedback on the five modifiable RFM items showed that the original binary Type 1/Type 2 response format did not adequately reflect all RC models. Panellists noted that some colleges combined both types within the same item, such as being partly community-based and partly located in a dedicated RC space, or offering course content both distinct from and integrated with clinical provision. Therefore, a third response option, \u0026ldquo;Both Type 1 and Type 2 (blended)\u0026rdquo;, was added for all modifiable items before Round 2.\u003c/p\u003e\n\u003cp\u003eSupplementary File 4 presents the wording changes made for Round 2 in response to Round 1 input. Items that had already reached consensus as culturally not difficult (i.e., RFM 2 Learning, 5 Social connectedness, and 7 Commitment to recovery) were retained largely unchanged; where qualitative comments informed revision, these changes were minor and intended only to clarify wording or improve cross-cultural accessibility.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRound 2 evaluation of revised wording\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn Round 2, all 11 revised items reached consensus for both understandability and cultural appropriateness (Table 4). For understandability, 73.2%\u0026ndash;95.1% of panellists rated the revised wording in the 4\u0026ndash;5 category, and for cultural appropriateness the corresponding range was 80.5%\u0026ndash;95.1%. These findings supported the acceptability of the revised wording across settings.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eTable 4. Round 2 quantitative findings\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eRevised item\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eUnderstandable in context\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e4\u0026ndash;5 n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eMedian (IQR)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eConsensus\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eCulturally appropriate\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e4\u0026ndash;5 n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eMedian (IQR)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eConsensus\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"7\"\u003e\n \u003cp\u003e\u003cstrong\u003eRCM mechanisms\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eEmpowering, autonomy-supporting environment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e30 (73.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e4 (3\u0026ndash;4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e34 (82.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e4 (4\u0026ndash;4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMoving toward an equal partnership\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e39 (95.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e38 (92.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eConnecting with peers and working together\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e37 (90.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e37 (90.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eFacilitating recovery\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e32 (78.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e4 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e33 (80.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e4 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"7\"\u003e\n \u003cp\u003e\u003cstrong\u003eRFM non-modifiable items\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e1. Valuing equality\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e39 (95.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e4 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e37 (90.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e2. Learning\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e37 (90.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e4 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e36 (87.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e3. Tailored to the student\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e36 (87.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e4 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e36 (87.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e4 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e4. Co-production of the RC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e33 (80.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e4 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e34 (82.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e4 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e5. Social connectedness\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e39 (95.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e4 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e39 (95.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e6. Community focus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e39 (95.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e4 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e35 (85.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e7. Commitment to recovery\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e39 (95.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e39 (95.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5 (4\u0026ndash;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eConsensus was predefined as \u0026ge;70% of responses in the 4\u0026ndash;5 category.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFinalised cross-cultural version of the RCM and RFM items\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn Round 2, all items reached consensus as culturally appropriate. Qualitative comments (1,157 words) were limited and largely confirmatory. Three minor wording refinements were implemented by the research team (Supplementary File 5). The final post-Round 2 items are presented as the Global RCM Version 1 (Global RCM v1) and Global RFM Version 1 (Global RFM v1), alongside the original England-developed wording, in Supplementary File 6.\u0026nbsp;\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis two-round international Delphi study used a cross-cultural de-centring approach, informed by corpus-based linguistic analysis of panellist feedback, to examine the cross-cultural applicability of the RCM and RFM. We found that cultural influence operated less at the level of whether RC principles were valued, and more at the level of how those principles were understood, worded and enacted. Core principles were broadly endorsed across countries, but concepts embedded in England-developed RC tools and wider high-income Anglophone recovery-oriented discourse, such as coproduction, shifting power and recovery itself, were not always directly applicable across settings without clarification. Cross-cultural de-centring helped make implicit assumptions explicit, distinguish personal and social recovery from clinical recovery, and reframe ideas such as tailoring and partnership in ways that were meaningful and feasible across diverse contexts. This is timely in light of current international emphasis on person-centred, rights-based and community-based mental health care.\u003csup\u003e15\u003c/sup\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eA central finding was that the underlying RC principles travelled better than the original English wording used to express them. In Round 1, all four RCM mechanisms and six of seven non-modifiable RFM items reached consensus for importance, yet consensus on low cultural difficulty was much more limited. This pattern suggests that panellists across settings broadly endorsed the constructs, but not always the phrasing through which those constructs had been operationalised. In cross-cultural psychiatry terms, the challenge was less about conceptual rejection than about semantic portability and contextual intelligibility.\u003csup\u003e2\u003c/sup\u003e The study therefore supports a distinction between valuing a principle and recognising it in a linguistic form shaped by a particular cultural and service discourse. This distinction is especially important for recovery-oriented frameworks, where apparently familiar terms may carry different institutional, clinical and moral meanings across settings.\u003csup\u003e16\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eThe study also makes visible how tools developed in high-income Anglophone settings can contain assumptions that are locally intelligible but not automatically universal. Terms such as “coproduction”, “shifting the balance of power” and “recovery” are well established within UK recovery-oriented policy and service reform discourse, and can therefore appear self-explanatory to researchers and practitioners working within that context. Internationally, however, these terms were not always interpreted consistently and sometimes carried unintended meanings.\u003csup\u003e17,18\u003c/sup\u003e For example, “shifting the balance of power” assumes a discourse of explicit power redistribution that may resonate in Anglophone service reform language, but may sound confrontational, abstract or structurally unclear elsewhere.\u003csup\u003e19\u003c/sup\u003e Likewise, although coproduction is now widely used in UK health policy and personalised care discourse, it does not travel as a universally shared term without explanation.\u003csup\u003e20\u003c/sup\u003e One interpretation is that this wording reflected assumptions common in some Anglophone service cultures, including relatively explicit egalitarian language, individual autonomy and tolerance of open-ended participation.\u003csup\u003e21,22\u003c/sup\u003e Across countries, panellists repeatedly sought clarification of meaning, role and relationship, suggesting that these assumptions were not consistently shared internationally. De-centring therefore did not replace the underlying RC principles; rather, it made their intended meaning more explicit and contextually interpretable for broader international use. This aligns with wider concerns in cross-cultural psychiatry about dominant systems mistaking local categories for universal ones.\u003csup\u003e23\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eThe item revisions illustrate how this de-centring process worked in practice. “Empowering environment” became “Empowering, autonomy-supporting environment”, making explicit a meaning that had been too dependent on English-language recovery vocabulary. “Shifting the balance of power” became “Moving toward an equal partnership”, replacing an abstract and potentially adversarial formulation with wording that was more relational and consistent with shared decision-making language. “Enabling different relationships” became “Connecting and working together”, removing ambiguity around what “different” meant. “Facilitating personal growth” became “Facilitating recovery”, avoiding a potentially culturally loaded assumption that improvement must take the form of upward growth. Similar refinements occurred within the Fidelity Measure. “Professionals” became “staff”, reducing dependence on systems in which “professional” implies formal registration or status. “Tailored to the student” was reframed to emphasise flexibility, reasonable adjustments and feasibility within local context and resources, rather than assuming a uniformly well-resourced service infrastructure. “Commitment to recovery” was expanded to specify that recovery refers to personal and social recovery, not only clinical recovery or symptom reduction. Collectively, these changes moved the framework away from locally saturated shorthand and toward wording that was more explicit, relational and transferable.\u003c/p\u003e\n\u003cp\u003eThese findings have implications beyond RCs. Mental health tools developed in high-income Anglophone contexts may be presented as broadly applicable while still embedding culturally and institutionally specific assumptions about service organisation, professional identity, rights discourse and interpersonal style. This is a general risk when any national or regional model becomes influential internationally. The present study shows why explicit de-centring is needed before such tools are adopted globally. For cross-cultural psychiatry, this supports a move away from assuming that transportability follows automatically from translation, and towards testing conceptual equivalence, semantic clarity and implementation feasibility across settings.\u003csup\u003e2\u003c/sup\u003e It also speaks to current debates on racialised and culturally unequal mental healthcare, where seemingly neutral service language can obscure the standpoint from which it was produced.\u003csup\u003e24\u003c/sup\u003e De-centring therefore has both methodological and ethical value: it improves interpretability while reducing the risk that culture-specific assumptions are universalised.\u003c/p\u003e\n\u003cp\u003eThe revised wording provides the Global RCM v1 and Global RFM v1 for research, service development and dialogue across international RC settings. These v1 tools are not intended to replace the original England-developed versions, but to retain interoperability with them while supporting more consistent interpretation across countries. This matters because international comparison requires enough continuity to compare findings, but enough de-centring to avoid imposing source-context assumptions uncritically. The Global RCM v1 and Global RFM v1 may therefore support cross-national RC research, local adaptation work, and policy dialogue in countries seeking to build recovery-oriented services without simply importing Anglophone terminology. In this sense, the revised tools function as resources for international learning rather than one-way model transfer.\u003c/p\u003e\n\u003cp\u003eThis study has several strengths. Its originality lies in applying an explicit cross-cultural de-centring approach to internationally used RC tools that had previously remained anchored in an English conceptual frame. Many cross-cultural mental health studies adopt frontal comparison,\u003csup\u003e25\u003c/sup\u003e in which tools or interventions developed in one source context are assessed against one or more target contexts. This can be useful for identifying adaptation needs, but may also reproduce a source-versus-target logic in which non-Anglophone settings are mainly understood by how they differ from an Anglophone reference point. By contrast, this study enabled a more inclusive lateral comparison across multiple RC settings, examining how concepts were interpreted from multiple standpoints and generating findings with broader international relevance. Its rigour lies in the mixed-method design: an international Delphi across 20 countries, predefined quantitative consensus thresholds, corpus-informed linguistic analysis of free-text responses, and iterative research team consultation to ensure that revisions were both evidence-informed and conceptually coherent. Its impact potential lies in producing the Global RCM v1 and Global RFM v1, which can be used more credibly in international RC research and service development while retaining interoperability with the England-developed RCM and RFM, allowing comparison between studies using either version.\u003c/p\u003e\n\u003cp\u003eThe study also has limitations. The sample was international but necessarily limited to countries where RCs exist or where relevant expertise was available; therefore, the study should not be interpreted as representing all world regions. However, given that RCs have been identified in 31 countries, participation from 20 countries represents substantial coverage of the current international RC community. Participation was still uneven across regions and countries, and some RC settings were represented by more panellists than others. The consultation was conducted in English, which may have favoured panellists more comfortable engaging with English-language recovery discourse, although this also made visible some of the linguistic assumptions that the study sought to examine. Round 2 established consensus on understandability and cultural appropriateness of the revised wording, but did not test translated versions or implementation in each context. Finally, because panellists were people already engaged with RCs, the study does not capture views from all stakeholders who may encounter these tools in wider policy or service environments.\u003c/p\u003e\n\u003cp\u003eFuture work should test how the Global RCM v1 and Global RFM v1 perform in translated versions and specific national settings, including whether further adaptation is needed at the level of examples, scoring or guidance for interpretation. Comparative work could also examine whether some items continue to function differently across linguistic regions, service systems or legal-political contexts. More broadly, the study supports a research agenda in cross-cultural psychiatry that treats tool adaptation as a substantive epistemic task rather than a final technical step. In conclusion, this Delphi study suggests that Recovery College principles are broadly valued internationally, but that wording developed in a high-income Anglophone context required de-centring before it could become more globally usable. The Global RCM v1 and Global RFM v1 therefore represent not a departure from the Recovery College model, but a clearer and more internationally applicable articulation of it.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eThis study followed the ACcurate COnsensus Reporting Document (ACCORD) reporting guideline for consensus methods (Supplementary File 2).\u003csup\u003e26\u003c/sup\u003e A study protocol was published in advance,\u003csup\u003e27\u003c/sup\u003e and the study materials and anonymised underlying data are openly available on the Open Science Framework (OSF; http://doi.org/10.17605/OSF.IO/CUAES). Ethical approval was obtained from the King’s College London Research Ethics Psychiatry, Nursing and Midwifery Subcommittee (reference: MRM-24/25-47085).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy design\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe conducted a two-round international Delphi study to examine the cross-cultural applicability of the RCM and RFM. The study followed Belton’s six-step Delphi methodology\u003csup\u003e28\u003c/sup\u003e and used a decentring approach to cross-cultural research,\u003csup\u003e29\u003c/sup\u003e which seeks to review and refine tools developed in one cultural context to improve their relevance across others. The Delphi aimed to identify: (1) the perceived importance of each RCM mechanism and RFM non-modifiable item, (2) the perceived cultural difficulty of achieving them, (3) recommended wording changes to improve cross-cultural applicability, and (4) which response category for each RFM modifiable item best aligned with local cultural context.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eParticipants and recruitment\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eParticipants were recruited through the RECOLLECT International Research Consortium,\u003csup\u003e30\u003c/sup\u003e which, at the time of the study, spanned 31 countries across Africa, Asia, Europe, North America, South America and Oceania. Country leads were invited to participate as panellists and to recruit additional eligible panellists within their national networks.\u003c/p\u003e\n\u003cp\u003eEligibility criteria were: age 18 years or older and experience of a RC in any role. To avoid over-representation from individual countries, recruitment was monitored by country, and country leads were asked to stop recruitment once five responses had been received, although any additional responses already submitted were retained. Consent was implied through survey completion, in line with standard Delphi practice.\u003csup\u003e31\u003c/sup\u003e Panellist anonymity was maintained through an individual online survey design, without access to the identities or responses of other panellists. This reduced the risk that panellists, particularly those with less experience or from countries where RCs are still emerging, might feel inhibited in expressing their views.\u003csup\u003e28\u003c/sup\u003e Names were given optionally for acknowledgement purposes and were accessible only to the Delphi administrator (YK), who had no voting rights in the consensus process.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSurvey development and procedure\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe draft survey was created by YK, reviewed by the research team, and piloted by country leads from Brazil, Estonia, Germany, Japan and Thailand to improve clarity, linguistic accessibility, suitability for online completion, and cross-cultural relevance. Revisions following piloting included adding clarifying examples to potentially ambiguous terms, and incorporating optional free-text comment boxes alongside quantitative responses.\u003c/p\u003e\n\u003cp\u003eThe finalised survey was used in Round 1 in English. Panellists rated the four RCM mechanisms and seven RFM non-modifiable items for (a) importance and (b) cultural difficulty of achieving them in their context. Rating used five-point Likert scales (1=”Not at all” to 5=”Very important/difficult”), and for each item panellists could also suggest wording changes to improve cultural applicability. Panellists then indicated for each of the five RFM modifiable items which of the two response categories better aligned with the cultural context of RCs in their country, and could provide wording suggestions. Round 1 remained open from July to November 2025 (16 weeks), closing after seven consecutive days without a new response.\u003c/p\u003e\n\u003cp\u003eThe Round 1 survey was then modified as described in the next section, to create the Round 2 survey. Round 2 remained open from January to February 2026 (5 weeks). Fortnightly reminders were sent to support retention. No financial incentives were used, as these may undermine the quality of panellist input in Delphi studies.\u003csup\u003e28\u003c/sup\u003e Panellists were offered social recognition through acknowledgement in the final paper.\u003csup\u003e32\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAnalysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor quantitative importance and cultural difficulty ratings, consensus was predefined as 70% or more of responses falling within the same category after collapsing the original five-point scale into three categories: responses of 1 and 2 were combined, 3 was retained as a middle category, and 4 and 5 were combined. This prespecified approach was used to reduce potential cross-cultural response bias, particularly variation in the use of extreme response options across cultural groups, and was supported by pilot findings indicating that a three-point structure would preserve meaningful distinctions while improving comparability across cultural groups.\u003csup\u003e27\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eDescriptive statistics were used to summarise responses, including frequencies, proportions, and medians (interquartile ranges (IQRs)). For RFM modifiable items, frequencies and proportions were calculated for Type 1 versus Type 2 responses. Items not reaching consensus in Round 1, or generating substantial qualitative concern, were revised and re-presented in Round 2.\u003c/p\u003e\n\u003cp\u003eFree-text responses in Round 1 were analysed using corpus-informed linguistic analysis, a software-assisted approach for examining large collections of digitised text, to identify recurrent terms and themes, patterns of language use, and ambiguities that informed wording changes to improve cross-cultural applicability. This approach differed from the published protocol,\u003csup\u003e27\u003c/sup\u003e which had proposed descriptive content analysis, because the responses consisted of brief, item-specific comments from panellists across 20 countries, making a systematic comparison of wording patterns especially appropriate. Open-text responses were grouped into datasets for each question type: cultural difficulty, components and mechanisms; importance, components and mechanisms; and cultural fit and wording. Keywords, defined as words statistically salient in one dataset compared with a reference dataset, were retrieved for each response dataset using the enTenTen21 corpus as reference (~52 billion words, available in Sketch Engine, the software used for the analysis). Salient themes and insights into language use and preferences were identified inductively by examining concordance lines, that is, strings of text displaying terms of interest in context. Findings were reviewed by the research team, including the developers of the RCM and RFM, and were used to generate revised wording for Round 2. In Round 2, panellists were asked whether the proposed revisions were appropriate in their contexts and, if not, to suggest further changes. Suggested revisions were reviewed and integrated by YK in discussion with co-authors. All panellists were sent the finalised Global RCM v1 and Global RFM v1 after Round 2.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthors’ contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualisation: YK; Methodology: YK, SVL, VL, CH, MS; Validation: YK, SVL, DN, RH, PL, JO, MCRA, YM, UB, JPP, VR, MN, TJ, VL, ST, AR, VK, AK, PB, CH, MS; Investigation: YK, SVL; Writing – original draft: YK, SVL; Writing – review \u0026amp; editing: YK, SVL, DN, RH, PL, JO, MCRA, YM, UB, JPP, VR, MN, TJ, VL, ST, AR, VK, AK, PB, CH, MS; Funding acquisition: CH, MS.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWe thank Marijn Antens, Simone Arbour, Ana Aznar Martínez, Anders Thorkil Bechgaard, Anne Mette Billekop, Lily Chan, Danielle Coppleson, Sally Easton, Matthew Jonathan Ellis, Daniela Fischer, Vinuri Fernando, Andrew Foster, Kumpon Kawkome, Kenji Kameda, Megumi Kurokami, Aya Kuroda, Ossi Kukkamäki, Ben Hart, Christine Larsen, Klaudia Lepska, Pernilla Lemvall, Living Recovery Center (Thailand), Sue Macdonald, Chiharu Nakayama, Andrew Nicholls, Jared Omundo, Apirak Paenmunin, Jörn Petersen, Supatchakarn Piros, Geoff Turner, Flavio Trolese, Paola Santo, Gianpaolo Scarsato, Erica Wan, Louisa Whorrod, Alex Thomas, Aurelie Tinland, Jonna Tolonen, Wilasinee Pinyowit, Kruawon Tiengtom, Asuka Yoshimi and Yoshie Morita for their support with this study.\u0026nbsp;\u003c/strong\u003eMS acknowledges the support of NIHR Nottingham Biomedical Research Centre.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study is funded by the National Institute for Health and Care Research (Recovery Colleges Characterisation and Testing (RECOLLECT) 2, ID: NIHR200605). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePatient consent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEthical approval was obtained from the King’s College London Research Ethics Psychiatry, Nursing and Midwifery Subcommittee (reference: MRM-24/25-47085).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStudy materials and underlying data are openly available on the Open Science Framework: http://doi.org/10.17605/OSF.IO/CUAES.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eWhitley, R., Shepherd, G. \u0026amp; Slade, M. Recovery colleges as a mental health innovation. \u003cem\u003eWorld psychiatry : official journal of the World Psychiatric Association (WPA)\u003c/em\u003e\u003cstrong\u003e18\u003c/strong\u003e, 141-142, doi:https://dx.doi.org/10.1002/wps.20620 (2019).\u003c/li\u003e\n\u003cli\u003eHenderson, C.\u003cem\u003e et al.\u003c/em\u003e Social inclusion of people with severe mental illness: a review of current practices, evidence and unmet needs, and future directions. \u003cem\u003eWorld Psychiatry\u003c/em\u003e\u003cstrong\u003e25\u003c/strong\u003e, 56-82, doi:https://doi.org/10.1002/wps.70031 (2026).\u003c/li\u003e\n\u003cli\u003eHayes, D.\u003cem\u003e et al.\u003c/em\u003e Organisational and student characteristics, fidelity, funding models, and unit costs of recovery colleges in 28 countries: a cross-sectional survey. \u003cem\u003eThe lancet. Psychiatry\u003c/em\u003e\u003cstrong\u003e10\u003c/strong\u003e, 768-779, doi:https://dx.doi.org/10.1016/S2215-0366(23)00229-8 (2023).\u003c/li\u003e\n\u003cli\u003eLin, E.\u003cem\u003e et al.\u003c/em\u003e Developing an evaluation framework for assessing the impact of recovery colleges: protocol for a participatory stakeholder engagement process and cocreated scoping review. \u003cem\u003eBMJ open\u003c/em\u003e\u003cstrong\u003e12\u003c/strong\u003e, e055289, doi:https://dx.doi.org/10.1136/bmjopen-2021-055289 (2022).\u003c/li\u003e\n\u003cli\u003eTheriault, J., Lord, M. M., Briand, C., Piat, M. \u0026amp; Meddings, S. Recovery colleges after a decade of research: a literature review. \u003cem\u003ePsychiatric Services\u003c/em\u003e\u003cstrong\u003e71\u003c/strong\u003e, 928-940, doi:10.1176/appi.ps.201900352 (2020).\u003c/li\u003e\n\u003cli\u003eCrowther, A.\u003cem\u003e et al.\u003c/em\u003e The impact of Recovery Colleges on mental health staff, services and society. \u003cem\u003eEpidemiology and psychiatric sciences\u003c/em\u003e\u003cstrong\u003e28\u003c/strong\u003e, 481-488, doi:https://dx.doi.org/10.1017/S204579601800063X (2019).\u003c/li\u003e\n\u003cli\u003eRonaldson, A.\u003cem\u003e et al.\u003c/em\u003e The impact of Recovery College enrolment on health service use and patient outcomes: retrospective matched cohort study using routinely collected data. \u003cem\u003emedRxiv\u003c/em\u003e, 2025.2012.2009.25341905, doi:10.64898/2025.12.09.25341905 (2025).\u003c/li\u003e\n\u003cli\u003eDeloitte. \u003cem\u003eThe shift to prevention: Realising the socio-economic potential\u003c/em\u003e, \u0026lt;https://www.deloitte.com/uk/en/Industries/life-sciences-health-care/research/the-shift-to-prevention-realising-the-socio-economic-potential.html\u0026gt; (2025).\u003c/li\u003e\n\u003cli\u003eHayes, D.\u003cem\u003e et al.\u003c/em\u003e Recovery Colleges Characterisation and Testing in England (RECOLLECT): rationale and protocol. \u003cem\u003eBMC Psychiatry\u003c/em\u003e\u003cstrong\u003e22\u003c/strong\u003e, doi:10.1186/s12888-022-04253-y (2022).\u003c/li\u003e\n\u003cli\u003e Toney, R.\u003cem\u003e et al.\u003c/em\u003e Development and Evaluation of a Recovery College Fidelity Measure. \u003cem\u003eCanadian journal of psychiatry. Revue canadienne de psychiatrie\u003c/em\u003e\u003cstrong\u003e64\u003c/strong\u003e, 405-414, doi:https://dx.doi.org/10.1177/0706743718815893 (2019).\u003c/li\u003e\n\u003cli\u003e Hayes, D.\u003cem\u003e et al.\u003c/em\u003e Evidence-based Recovery Colleges: developing a typology based on organisational characteristics, fidelity and funding. \u003cem\u003eSocial Psychiatry and Psychiatric Epidemiology\u003c/em\u003e\u003cstrong\u003e59\u003c/strong\u003e, 759-768, doi:10.1007/s00127-023-02452-w (2024).\u003c/li\u003e\n\u003cli\u003e Kotera, Y.\u003cem\u003e et al.\u003c/em\u003e Cross-Cultural Insights from Two Global Mental Health Studies: Self-Enhancement and Ingroup Biases. \u003cem\u003eInternational Journal of Mental Health and Addiction\u003c/em\u003e\u003cstrong\u003e23\u003c/strong\u003e, 3572-3581, doi:10.1007/s11469-024-01307-y (2025).\u003c/li\u003e\n\u003cli\u003e Griner, D. \u0026amp; Smith, T. B. Culturally adapted mental health intervention: A meta-analytic review. \u003cem\u003ePsychotherapy (Chic)\u003c/em\u003e\u003cstrong\u003e43\u003c/strong\u003e, 531-548, doi:10.1037/0033-3204.43.4.531 (2006).\u003c/li\u003e\n\u003cli\u003e Benish, S. G., Quintana, S. \u0026amp; Wampold, B. E. Culturally adapted psychotherapy and the legitimacy of myth: a direct-comparison meta-analysis. \u003cem\u003eJ Couns Psychol\u003c/em\u003e\u003cstrong\u003e58\u003c/strong\u003e, 279-289, doi:10.1037/a0023626 (2011).\u003c/li\u003e\n\u003cli\u003e Sj\u0026ouml;berg, A., Liv, P. \u0026amp; Lindstr\u0026ouml;m, M. Effect of Everyday Life Rehabilitation on recovering quality of life in individuals with serious mental illness in supported accommodation: a pragmatic cluster randomised controlled trial. \u003cem\u003eBMJ Mental Health\u003c/em\u003e\u003cstrong\u003e28\u003c/strong\u003e, e301757, doi:10.1136/bmjment-2025-301757 (2025).\u003c/li\u003e\n\u003cli\u003e Kuek, J. H. L., Raeburn, T. \u0026amp; Wand, T. Asian perspectives on personal recovery in mental health: a scoping review. \u003cem\u003eJ Ment Health\u003c/em\u003e\u003cstrong\u003e32\u003c/strong\u003e, 517-533, doi:10.1080/09638237.2020.1818709 (2023).\u003c/li\u003e\n\u003cli\u003e Kotera, Y.\u003cem\u003e et al.\u003c/em\u003e Cross-cultural Comparison of Recovery College Implementation Between Japan and England: Corpus-based Discourse Analysis. \u003cem\u003eInternational Journal of Mental Health and Addiction\u003c/em\u003e\u003cstrong\u003e23\u003c/strong\u003e, 4362-4386, doi:10.1007/s11469-024-01356-3 (2025).\u003c/li\u003e\n\u003cli\u003e Kotera, Y.\u003cem\u003e et al.\u003c/em\u003e Recovery College Advertising in the Netherlands: A Dutch-English-Japanese Corpus-Based Comparison and Its Implications for Quality-of-Life Orientation. \u003cem\u003e[preprint]\u003c/em\u003e, doi:10.1101/2025.08.13.25333552 (2025).\u003c/li\u003e\n\u003cli\u003e Kotera, Y.\u003cem\u003e et al.\u003c/em\u003e 28-country global study on associations between cultural characteristics and Recovery College fidelity. \u003cem\u003enpj Mental Health Research\u003c/em\u003e\u003cstrong\u003e3\u003c/strong\u003e, doi:10.1038/s44184-024-00092-9 (2024).\u003c/li\u003e\n\u003cli\u003e Kotera, Y.\u003cem\u003e et al.\u003c/em\u003e Cultural influences on fidelity components in recovery colleges: a study across 28 countries and territories. \u003cem\u003eGeneral Psychiatry\u003c/em\u003e\u003cstrong\u003e38\u003c/strong\u003e, e102010, doi:https://dx.doi.org/10.1136/gpsych-2024-102010 (2025).\u003c/li\u003e\n\u003cli\u003e Hofstede, G., Hofstede, G. J. \u0026amp; Minkov, M. \u003cem\u003eCultures and organizations: Software of the mind\u003c/em\u003e. 3rd edn, (McGraw-Hill Education, 2010).\u003c/li\u003e\n\u003cli\u003e Nazarian, A., Ranjbaran, A., Afifi, M., Atkinson, P. \u0026amp; Sadeghilar, N. Cross-cultural examination of effects of ethical leadership of independent hotels based on Social Exchange Theory. \u003cem\u003eInternational Journal of Cross Cultural Management\u003c/em\u003e\u003cstrong\u003e24\u003c/strong\u003e, 463-484, doi:10.1177/14705958241270817 (2024).\u003c/li\u003e\n\u003cli\u003e \u003cem\u003eTextbook of Cultural Psychiatry\u003c/em\u003e. 2 edn, (Cambridge University Press, 2018).\u003c/li\u003e\n\u003cli\u003e Stepanova, E.\u003cem\u003e et al.\u003c/em\u003e \"I am not a priority\": ethnic minority experiences of navigating mental health support and the need for culturally sensitive services during and beyond the pandemic. \u003cem\u003eBMJ Mental Health\u003c/em\u003e\u003cstrong\u003e28\u003c/strong\u003e, e301481, doi:10.1136/bmjment-2024-301481 (2025).\u003c/li\u003e\n\u003cli\u003e Schnegg, M. \u0026amp; Lowe, E. D. \u003cem\u003eComparing Cultures: Innovations in Comparative Ethnography\u003c/em\u003e. (Cambridge University Press, 2020).\u003c/li\u003e\n\u003cli\u003e Gattrell, W. T.\u003cem\u003e et al.\u003c/em\u003e ACCORD (ACcurate COnsensus Reporting Document): A reporting guideline for consensus methods in biomedicine developed via a modified Delphi. \u003cem\u003ePLoS Med\u003c/em\u003e\u003cstrong\u003e21\u003c/strong\u003e, e1004326, doi:10.1371/journal.pmed.1004326 (2024).\u003c/li\u003e\n\u003cli\u003e Kotera, Y.\u003cem\u003e et al.\u003c/em\u003e Enhancing cross-cultural applicability in recovery colleges: A global Delphi study protocol. \u003cem\u003ePloS one\u003c/em\u003e\u003cstrong\u003e20\u003c/strong\u003e, e0332729, doi:https://dx.doi.org/10.1371/journal.pone.0332729 (2025).\u003c/li\u003e\n\u003cli\u003e Belton, I., MacDonald, A., Wright, G. \u0026amp; Hamlin, I. Improving the practical application of the Delphi method in group-based judgment: A six-step prescription for a well-founded and defensible process. \u003cem\u003eTechnological Forecasting and Social Change\u003c/em\u003e\u003cstrong\u003e147\u003c/strong\u003e, 72-82, doi:https://doi.org/10.1016/j.techfore.2019.07.002 (2019).\u003c/li\u003e\n\u003cli\u003e van de Vijver, F. J. R. \u0026amp; Leung, K. \u003cem\u003eMethods and Data Analysis for Cross-Cultural Research\u003c/em\u003e. 2 edn, (Cambridge University Press, 2021).\u003c/li\u003e\n\u003cli\u003e Recovery Research Team. \u003cem\u003eRECOLLECT International Research Consortium (RIRC)\u003c/em\u003e, \u0026lt;https://www.researchintorecovery.com/recollect-international-research-consortium-rirc/\u0026gt; (2025).\u003c/li\u003e\n\u003cli\u003e Toma, C. L. \u0026amp; Picioreanu, I. The Delphi Technique: Methodological Considerations and the Need for Reporting Guidelines in Medical Journals. \u003cem\u003eInternational journal of public health research\u003c/em\u003e\u003cstrong\u003e4\u003c/strong\u003e, 47 (2016).\u003c/li\u003e\n\u003cli\u003e Khodyakov, D., Grant, S., Kroger, J. \u0026amp; Bauman, M. (RAND Corporation, Santa Monica, CA, USA, 2023).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-9614758/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9614758/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Recovery Colleges are expanding internationally (264 identified in 31 countries across six continents), yet key tools for defining and assessing them were developed in high-income Anglophone contexts. We conducted a two-round international Delphi study using a cross-cultural de-centring approach to examine the cross-cultural applicability of the RECOLLECT Change Model and RECOLLECT Fidelity Measure, and to develop globally applicable and interoperable versions of these tools. In Round 1, 54 panellists from 20 countries rated item importance and cultural difficulty and provided free-text feedback, analysed using corpus-informed linguistic methods. Core principles were broadly endorsed, but “coproduction”, “recovery”, “professionals” and “shifting the balance of power” were interpreted inconsistently, revealing embedded assumptions about equality, autonomy and open-ended participation. Round 1 findings informed wording refinements tested in Round 2 with 41 panellists from 15 countries. All 11 revised items reached consensus for understandability and cultural appropriateness, strengthening the tools while preserving core Recovery College values.","manuscriptTitle":"Cross-cultural de-centring of Recovery College tools: a 20-country Delphi study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-06 05:20:46","doi":"10.21203/rs.3.rs-9614758/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"353ae34c-42fc-467d-9e16-79bb65bc1167","owner":[],"postedDate":"May 6th, 2026","published":true,"recentEditorialEvents":[{"type":"submitted","content":"Nature Mental Health","date":"2026-05-05T06:45:56+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":67528702,"name":"Health sciences/Health care/Health services"},{"id":67528703,"name":"Health sciences/Health care/Patient education"}],"tags":[],"updatedAt":"2026-05-06T05:20:46+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-06 05:20:46","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9614758","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9614758","identity":"rs-9614758","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}