UPSIDES - Recovery Through Peer Support

Peer support (PS) is a well-established intervention within mental health services (MHS) internationally, involving a person recovering from severe mental health problems (SMHP) in supporting other users. International studies have shown that PS strengthen people with SMHP recovery and social inclusion but also highlight the need for larger effect studies on specific manualized PS interventions. Despite international research showing benefits of PS in MHS, PS an underused type of support in Sweden. There is a clear and urgent need for more Swedish research developing and evaluating the efficacy of PS interventions for persons recovering from SMHP in terms of their recovery journey) and in shifting Swedish mental health services towards a recovery focussed model. In meeting this need the current project aims to evaluate the efficacy of a novel co-produced and manualized PS intervention, the UPSIDES-Swe, in Swedish MHS. Our objectives of introducing and evaluating this peer support intervention aims at increasing the crucial aspects of both recovery, social inclusion and related health aspects for the growing group of people with SMHP, while also investigating implementation issues of a new intervention. The prevalence of people living with SMHP in Sweden is increasing and there has long been a call for the evidence-based interventions in MHS aiming to improve the recovery of this vulnerable group. Historically, mental health services have almost exclusively focussed on delivering interventions and strategies that stress users compliance with treatment plans, largely failing to consider an individual's personal recovery goals or self-determination. In contrast, recovery oriented services reject this paternalistic approach, delivering support with users as active partners in their care. At the core of the recovery model is the principle that people in recovery from SMHP are, by virtue of their lived experience, uniquely qualified to support other users within MHS. Peer support workers (PSW) are those with lived experience of mental illness who have experienced their own recovery process, and once trained are employed to provide support to others with SMHP. PS is hence an essential component of recovery-oriented services, underpinned by the belief that those having personally experienced mental illness and a recovery process, are uniquely positioned as equal partners to offer support, encouragement, and hope to others recovering from SMHP. Early international research on PS highlighted that PSW were central in enabling positive long-term outcomes in those with SMHP. For example, service users supported in their recovery journey by PSW reported better quality of life, recovery, empowerment and functioning, improved social functioning and social integration, increased experience of social inclusion, better self-esteem, a sense of hope and control, all essential components in recovering from a mental illness. A parallel line of research examining the efficacy of recovery models utilizing PSW suggests that service users required less support, reduced their substance abuse, had fewer hospitalizations, and experienced decreased levels of depression and psychosis. While there is a significant body of evidence supporting the use of the recovery model and PSW, there is a clear need for more rigorous and longitudinal research in informing an optimal approach to service delivery and the impact of this model on longitudinal outcomes. Also, a recent systematic review of randomized controlled trials (RCTs) investigating peer support interventions concluded that future research was needed utilising RCTs with robust designs and comparable co-created interventions to further investigate the evidence of peer support interventions. While initially introduced in the USA in the nineties, PSW are now internationally an essential and natural component of many CMHS but in Sweden there is still much work to be done. The PS movement in Sweden emerged in mid-2010 with the leading user organization in Sweden, The Swedish Partnership for Mental Health (in Swedish: Nationell Samverkan för Psykisk Hälsa) (NSPH) educating and implementing PSW in services. In developing the Swedish PS educational program and model, the NSPH has extensively collaborated with both researchers in the field, who are also the applicants of this grant, and leading international organizations implementing PSW in other countries. Results of the implementation of PSW in Sweden (our pre studies) demonstrated that while this support was key in users' recovery PS also needed further attention regarding both what type of interventions PSW provided and also regarding implementation issues due to a lingering medical paradigm in which expert knowledge is prioritized over the experiential. Even though users report being fully supported towards recovery by PSW, PSW is still an underused resource in the Swedish MHS workforce with national guidelines requesting further studies to increase evidence. Providing evidence for this workforce would also result in work opportunities as a PSW for the neglected group of people with SMHP. Therefore, there is an urgent need for research evaluating the implementation of a recovery-oriented PS intervention, specifically tailored towards the context of the Swedish mental health system, and its impact on outcomes at the individual and system levels. UPSIDES is a six-country research consortium with expert researchers within recovery and PS that aims to help replicate and scale up formal PS for people with SMHP. In cooperation with users, UPSIDES has developed the UPSIDEs intervention and is currently conducting a multi-site randomised controlled trial (RCT) to examine the effect of the UPSIDES peer support intervention in a geographically diverse selection of study sites. UPSIDES is a 12-week group intervention building on different themes focusing on building knowledge and skills crucial to the recovery journey such as understanding and sharing personal recovery experiences, developing personalized recovery plans, and identifying individual needs, building effective relationships with health professionals, gaining insights into the perspectives of relatives, and promoting social networking and community engagement. The UPSIDES model is that the project has followed its implementation in a diversity of welfare and health care systems. This has provided a collaboration with a unique insight into translating and adapting such a model during the implementation process. The current project is part of, and continues to build on, the overarching FORTE financed research program UserInvolve that in collaboration with NSPH investigates a variety of user involvement interventions and movements in Swedish CMHS. In addition, our project on PS has initiated a collaboration with the international research organization UPSIDES, in order to translate, co-produce and test the UPSIDES intervention in a Swedish context. To prepare for implementation and testing of such an intervention, necessary pre studies on the current status of PSW and types of PS support in Swedish settings have been conducted (submitted), an extensive process of co-designing the UPSIDES intervention with PSW, users and NSPH (in manuscript) and testing it in a pilot study (ongoing). This has increased the complexity and time frame of the project within the UserInvolve program, but has resulted in the unique co-produced UPSIDES-Swe, fit to be implemented in Swedish MHS.

In this first RCT in a Swedish setting, the overall aim is to investigate the efficacy of this co produced Swedish version of the group based and manualized peer support intervention, the USIDES-Swe, as compared to treatment as usual (TAU) among 300 service users. We hypothesize that the peer support intervention will be more effective than TAU in terms of recovery (primary outcome at the individual level) as assessed by Questionnarie of the Process of Recovery (QPR) (17), at intervention end. Secondary outcomes such as social inclusion, quality of life, empowerment, as well as symptoms and function will also be targeted and assumed to be in favour of the intervention group. A process evaluation will also be conducted. The specific aims are to evaluate the UPSIDES-Swe including studying the effects on recovery, social inclusion and the other health related factors mentioned, as well as discerning critical implementation factors through a process evaluation.

The research questions are:

Is the peer support intervention, UPSIDES-Swe, significantly more effective than TAU at intervention end (10 weeks) in terms of personal recovery (primary outcome)? Is UPSIDES-Swe significantly more effective than TAU post intervention and at three month followups in terms of other personal recovery aspects (secondary outcomes i.e., social inclusion, empowerment, quality of life)? Is the UPSIDES-Swe, significantly more effective than TAU post intervention and at three month followups in improving secondary outcomes of clinical recovery related variables, i.e., symptoms and functioning? How well can the implementation plan and UPSIDES-Swe intervention be implemented and processes explained in relation to primary and secondary outcomes?

Study design and methods Significant work has been put into the development phase according to Medical Research Council (MRC) framework, in building the knowledge base in the area of peer support. A number of preparatory studies have been conducted highlighting users´ positive experiences of receiving peer support, managers' views of having a PSW in the team, and mapping the current landscape of PSW in Sweden (submitted). The process has also supported the co-producing (with PSW and service users) of a manualized PSW intervention tailored to the Swedish context, called UPSIDE-Swe (unpublished data). The motivation for undertaking this research is driven by a call from service users and advocacy agencies for MHS to deliver more personalised, recovery oriented services. The next logical step in the MRC framework is to evaluate the efficacy of the PSW intervention, while also conducting a process evaluation. Co-production with user organization NSPH is ongoing in all part of the current project with further planned collaboration, as also stated in the signed agreement from NSPH (see attachment).

Research design of RCT A randomized, assessor-blinded, multi-center trial will be performed between the years 2025-2028. A total of 300 service users will be recruited, and half will be randomized to receiving a co-produced peer support intervention (UPSIDES-Swe), added to TAU, and an equal number to TAU. The main hypothesis is that participants in the intervention group will experience better personal recovery (according to QPR instrument) when compared to the group only receiving TAU directly after the intervention and at three month followups. Intention-to-treat (ITT) will be applied and to ensure the trial quality. In all, the trial will be in line with the CONSORT (Consolidated Standards Of Reporting Trials) statement for reporting parallel and non-pharmacological trials, see Figure 1. The protocol publication will be according to SPIRIT framework (Standard Protocol Items: Recommendations for Interventional Trials). The SPIRIT figure is used (Figure 2). A process evaluation will also be performed using qualitative and quantitative methods. Eligible participants must reside in Sweden and be assigned to a MHS. Other criteria are a) being over 18 of age and b) having SMHP (including a diagnosis of mood disorder, anxiety disorder, e.g., PTSD, panic disorder etc., and/or psychotic disorder, for a duration of >2 years). Potential participants will be excluded if they are acutely unwell or cognitively impaired to an extent that they cannot make an informed decision and/or participate in the group program. An employed PSW at units in community-based psychiatry will provide oral information and invite service users meeting the inclusion criteria to participate in the study. Those interested will be provided with written information about the trial and invited to meet with the PI and research colleagues to discuss the study further. Informed consent will be obtained from all participants agreeing to participate in the study.

Randomization and blinding The study is a single-blinded RCT. Participants will be randomized to the study groups following their baseline assessment by an administrator independent of the research team. All researchers will be blind to group allocation. All research data will be encrypted, generated and downloaded to the RedCap digital data management system and kept according to the safe data-secure LUSEC system of Lund University. While it is not possible to blind study participants (PSW and service users in both study groups), they will be instructed on the importance of not disclosing their group allocation to the research team.

Interventions

The UPSIDES-Swe This manualized group-based intervention was developed and co-produced within the multisite program UPSIDES and is currently being tested via RCTs in numerous countries. As part of the research program UserInvolve the intervention UPSIDES has been co-designed with stakeholders such as PSW, service users and NSPH, aligning the program with Swedish MHS and cultural context resulting in the unique UPSIDES-Swe (in manuscript). The description of the UPSIDES-Swe peer support program aligns with the TIDieR Checklist. The UPSIDES-Swe intervention incorporates key principles fundamental to peer support and central to the concept of personal recovery, focusing on individuals living satisfying and fulfilling lives beyond the limitations of mental illness. These values include hope and encouragement, focusing on individual resources for regaining autonomy and empowerment, and opening up to new opportunities. Originating from narrative therapy, the tree of life model is used in the UPSIDESSwe, and focuses on establishing a secure and supportive working environment, identifying each individual's strengths and resources, and developing trusting relationships with peers based on shared experiences. It is anticipated that UPSIDESSwe will be a catalyst to better personal recovery (primary outcome), social inclusion, quality of life, empowerment, and clinical recovery in forms of increased functioning and decreased symptoms. The model of service delivery involves two PSW, who have recovered from SMHP and undergone training on the recovery approach, and will facilitate face-to-face group sessions with groups of service users. PSW will receive a comprehensive manual and practical training on effectively facilitating UPSIDESSwe groups. UPSIDESSwe will be delivered in the CMHS contexts, with a suitable group room. Importantly, PSW will be supervised bi-weekly by NSPH staff (people with own lived experience) and experienced health professionals on a digital platform, enabling them to exchange experiences and obtain support. Additionally, NSPH staff and researchers will be available for online consultation and support as needed. Each UPSIDESSwe group will comprise of ten weekly group sessions, with between 4-8 participants, lasting between one to one and a half hours. Sessions are bookended by an opening and closing round, where participants 'check-in', sharing their thoughts and feelings. The opening round focuses on experiences of the previous week with the closing round summarizing the key messages from the session. The weekly sessions take the form of discussions and group activities, focusing on topics relevant to mental health, aligned with the needs and preferences of each group. UPSIDESSwe begins by initiating group facilitation, establishing the groundwork for delivering peer support in a group setting. Additionally, the emphasis on communication and listening is on cultivating active listening skills and employing recovery-oriented language. UPSIDESSwe focuses on building knowledge and skills crucial to the recovery journey. This includes understanding and sharing personal recovery experiences, developing personalized recovery plans, and identifying individual needs, supports, hopes, and dreams. Furthermore, the program incorporates building effective relationships with health professionals, gaining insights into the perspectives of relatives, and promoting social networking and community engagement. In examining fidelity, a scale (see Table 1) will be distributed to service users and PSW, encompassing dosage (the number, frequency, and duration of peer support sessions), as documented in regularly administered monitoring and evaluation (M\&E) forms, and implementation outcomes.

TAU Those allocated to the TAU group will receive CMHS support as usually provided by staff. Generally, TAU typically focuses on traditional treatment and includes a supportive contact with a health professional in the general MHS team with little integration of support from other professionals or services. Treatment in MHS, which service users may also be provided while being connected to MHS, may focus on forms of time-limited individual psychological treatment such as Cognitive Behaviour Therapy (CBT) or interventions such as Individual Placement and Support (IPS). Such traditional treatments do not generally focus on personal recovery or social inclusion of the service users. Recovery focused self-management interventions are scarce and the opportunities for specific group based and manualized peer support interventions does not exist. The project ensures consistency in the type of care each participant receives in TAU since all units with PS are known through our pre studies. Background questionnaires will also ask for current treatment in TAU, at all measure points, and the project will generally avoid collapsing with other potential developmental/research programs that may include any recovery-oriented efforts, such as those provided by the UserInvolve program. Also, in order to prevent contamination will education of clinicians and participants against contamination and the provision of clear information about the purposes of the trial be provided.

Outcomes and data collection Data collection will be conducted through online surveys (all instruments psychometrically tested) via RedCap. As presented in the table, the measures respond to research questions (hypotheses). The selection of instruments is based on the results from the preparatory studies and international research on peer support. Feedback on the measurement framework has also been sought from representatives from our partner organizations. In combination with this knowledge there is a clear line of measures that are of relevance to use in the current study. The measurement framework includes a measure of recovery (primary outcome), aspects of social inclusion, quality of life, empowerment and clinical recovery (secondary outcomes). The primary outcome, a questionnaire measuring the process of recovery is well validated and frequently used measurement in research and MHS. Secondary outcome measures of personal and clinical recovery are measures of health outcomes relevant to the target group as a whole. The impact of personal factors (gender, ethnicity and sociodemographic status) on outcomes will also be investigated. A socio-demographic and clinical questionnaire will obtain information on relevant factors. Data will be collected from both arms (UPSIDESSwe and TAU) of the study at baseline (0 weeks, prior to the intervention), following the intervention period (at 10 weeks), and at three months followups. The time frame adheres to earlier studies on peer support such as a recent Danish RCT study examining a peer support group intervention with the same primary measure and also our own earlier intervention studies using the QPR. Peer support is estimated to have a quite quick effect on recovery and hence are the time frames not as extensive, with three months as a followup measure. Prior testing of the measurement framework has shown that the battery of instruments takes between 30-40 minutes to complete. Information relating to adverse events will, with consent, be obtained from participants' medical records. See Table 1 for a full data collection overview of quantitative data.

Sample size and statistics With a critical alpha value of 0.05, a sample size of 300 people with an attrition rate of 20 %, i.e., N=240 will generate a power of 80 % to detect a standardized difference of d=0.37. Based on a study comprising of 335 participants, with a QPR15 score SD of 11.56 on the QPR scale ranging from 19-90, a sample mean difference of 5 scoring points would thereby be detectable. The principle of intention-to-treat (ITT) will be applied and the level of significance for all statistical tests will be .05, two-tailed. General linear models (GLM) will be used to analyze normally distributed data, including the primary outcome. For non-normally distributed data, non-parametric statistics will be used. Effect sizes will be calculated for completers and completers and non completers combined (ITT). Cronbach alpha will be calculated for internal consistency of all measures. A statistical unit will be consulted for calculations of trial results and may slightly alter the final analysis plan. Statistical analyses will be performed using the IBM SPSS statistical software package, and Repeated Measures ANOVA will be used to analyze differences between groups and over time.

Process evaluation The Medical Research Framework for complex interventions will guide the overall research process with consideration for those factors acting as barriers or facilitators in implementing the UPSIDESSwe program. This project will also undertake multiple interviews with PSW and their managers, enabling examination those factors impacting UPSIDESSwe implementation and understanding of the reality of implementing UPSIDESSwe with Swedish MHS (process evaluation), and investigation of the outcomes and experiences not captured by questionnaires. MSH managers, first-line managers and team staff will be interviewed at the beginning and at followups and three months after completion of the intervention period. Furthermore, twenty participating service users in the peer support intervention will be strategically selected for in-depth interviews to shed light on their experience of receiving peer support and to get a multifaceted image of the intervention and the informants' view on possible increased community integration and enhanced personal recovery. Individual and focus groups interviews will be used as well as a survey on the content and delivery of UPSIDES. Facilitators will also fill out the UPSIDES fidelity scale on peer support at end of intervention. See Table 2 for full overview of data collection for process evaluation according to MRC framework. The qualitative data of managers, staff PSW will be analyzed deductively using the MRC framework examining those factors acting as barriers of facilitators in implementing UPSIDES-Swe, to complement information to the RCT study. In depht interviews with service users who have gone through the intervention will be analyzed via content analysis in order to reveal any additional themes regarding for example mechanisms of impact.

Time plan and procedure The previous preparatory studies make the current project feasible, with the RCT being conducted over three years, from July 2025-June 2028. The first two years are dedicated to recruiting units (MHS), educating stakeholders and PSW, and conducting efficacy testing via the RCT (2025- 2027), with the final year focusing on analyzing and disseminating the research findings (2028). Related qualitative study on experiences of receiving and delivering the intervention will be performed within the study period. Based on our previous experience of multisite trials, recruitment may extend to six months. This means that although the intervention spans ten weeks for each study participant, the RCT as a whole with the number of participants required for sufficient study power may take up to two years. Prior to commencing the RCT, PSW at ten CMHS (sited in Skåne, Västra Götaland, Stockholm etc.) will be educated on the intervention. Skåne, Stockholm, Västra Götaland etc have a total of about a hundred PSW employed. Our ambition in the current project is to perform the trial simultaneously at all sites, i.e., interventions, data collections, fidelity measures are coordinated in time as much as possible.