What Is Mental Health Peer Support Work?

Mental health peer support (hereafter “PS”) is relationship-based work between mental health service users and PS workers (PSWs) (Kotera et al., 2022). PSWs are people with lived experience of using mental health services and/or of mental health problems. The shared experience of mental health problems between a service user and a PSW is central in PS and promotes mutuality and connection (White et al., 2020). PS aims to support service users on the issues they consider important to their own lives and recovery (World Health Organization, 2021).

Since the 1990s, PS has been available in many countries (Davidson et al., 2012). In 2006 it was estimated that there were over 10,000 PSWs in the USA (Goldstrom et al., 2006). In England, PS is embedded in the national mental health implementation plan, which projects more employment of PSWs (National Health Service England 2019). PS is person centred and is in line with the current global emphasis on advocating human rights (World Health Organization, 2021). Moreover, PS is cost-effective and can be implemented in countries across different socioeconomic levels (Pearson et al., 2021). For example, the UPSIDES (using peer support in developing empowering mental health services) PS intervention has been evaluated in a range of high-, middle-, and low-income countries (Germany, UK, Israel, India, Uganda, and Tanzania) (Moran et al., 2020).

Evidence Base

PS has a considerable evidence base (Mutschler et al., 2021). A meta-analysis synthesised evidence from 19 randomised controlled trials (RCTs) of one-to-one PS and reported that PS was effective for recovery, empowerment, and working alliances (White et al., 2020). Another meta-analysis, focused on eight group-based PS RCTs, found evidence for effectiveness in relation to recovery and psychiatric symptoms (Lyons et al., 2021). For people with severe mental illness, a meta-analysis found little evidence on hospitalisation, overall symptoms, or satisfaction with services; however, some evidence on hope, recovery, and empowerment was found (Lloyd-Evans et al., 2014). A systematic review of digital PS, in which PS was offered live or automated using technology, also identified that digital PS is feasible, acceptable, and effective (Fortuna et al., 2020).

Qualitative evidence also demonstrates PS effectiveness. A meta-synthesis of 27 qualitative studies reported that service users view their PSW as a role model, which supports service users’ hope and motivation for recovery (Walker & Bryant, 2013). Another meta-synthesis based on 34 studies found that PS enables service users to reframe their identity, facilitated by mutual relationships and story sharing by the PSW (MacLellan et al., 2015). A longitudinal study revealed that shared stories between a PSW and a service user help change fears into hope (Simmons et al., 2020).

Additionally, PS has been evaluated with diverse populations, including older adults, youth, LGBTQ+ people, persons with disabilities, and those in a forensic context (Borthwick et al., 2020; Shalaby & Agyapong, 2020; Williamson & Durcan, 2020). The adaptability of PS to many different groups is high (Charles et al., 2020).

Knowledge Gap

A few attempts have been made to identify mechanisms of how PS works (Evans, 2022; Watson, 2019). However, the active ingredients, hereafter “components” of PS are not clear. Components refer to “elements of an intervention that have the potential to causally influence outcomes” and are “directly related to an intervention theory of change” (Caldwell & Welton, 2020). Identifying components is important in understanding and improving a multi-component intervention (Petropoulou et al., 2021) such as PS. Specific knowledge gaps relate to what components are used in PS, the relative effectiveness of each component, and whether the overall impact of combining components is neutral (no benefit from adding multiple components), additive (each component provides independent benefit), or synergistic (a multiplicative effect exceeding the benefit from individual components). To address these knowledge gaps, a systematic review and component network meta-analysis (CNMA) are planned (PROSPERO: author blind). The planned analysis requires a typology of PS components to allow all administered PS to be coherently described and categorised.

A typology is an explanation and classification of complex interventions based on qualitative and/or quantitative analysis (Smithman et al., 2020). The purpose of a typology is to classify a whole phenomenon into discrete yet interrelated categories (Ayres & Knaft, 2008). For example, the Consolidated Framework for Implementation Research (CFIR) (Damschroder et al., 2009) is a typology and is the most used framework in implementation science (Damschroder et al., 2022). CFIR helps manage a complex process of implementation by identifying key areas that can impact the implementation success (Keith et al., 2017). PS is a complex, multi-component intervention, which can benefit from development of typologies.

Study Aim

The aim of this study was to develop a typology of components for one-to-one PS in adult mental health services and to thematically categorise them. Our findings will help improve the coherence and compatibility of future evaluation and implementation work in PS, and allow the conduct of a coherent and clinically relevant component meta-analyses.

Methods

Design

Systematised review and expert consultation, using an established six-step framework: (1) identify the research questions; (2) identify relevant studies; (3) study selection; (4) chart the data; (5) collate, summarise, and report the results; and (6) expert consultation (Arksey & O'Malley, 2005).

Procedures

A systematised review was conducted to create an initial list of components of one-to-one PS for adults in mental health services, comprising components and sub-components. A systematised review involves part of the process of a systematic review; however, it does not meet the rigour of a systematic review (Grant & Booth, 2009; Sataloff et al., 2021). We have made a decision to conduct a systematised review, because we were interested in what components were commonly present in PS research. We judged no additional components would be identified by conducting a full systematic review. We only included RCTs because (a) RCTs report rigorous evidence for the effectiveness (Hariton & Locascio, 2018), which would help identify components, and (b) there were enough RCTs of one-to-one adult service user PS (White et al., 2020). Three types of publication were used to create the PS initial component list.

Type 1: Trial Reports

To identify relevant RCTs, first, included publications in a recent systematic review of one-to-one PS for adults were full-text reviewed by YK (White et al., 2020). To find additional RCTs, Medline and PsycINFO searches were conducted in August 2022 by YK after a consultation with a subject librarian for health sciences (AA). The search strategy was developed focusing on the key elements: RCTs, peer support work, mental health, one to one (Electronic Supplementary Material 1). Titles and abstracts for the retrieved articles were screened by YK, after 10% screened independently by AC and concordance established. Full-text reviews were then performed by YK, and the included articles were compared with the eligibility criteria by YK and AC.

Type 2: PS Training Manuals

Published training manuals (including those formally published and those available online) were identified through online searching of PS websites.

Type 3: Published Change Models

Published change models in PS were identified through backward citation tracking of included publications to identify studies characterising the mechanism of PS.

Once a preliminary set of included documents from all three types of publications were collated, an expert consultation was conducted to maximise the sensitivity of the review. Experts were selected to comprise a wide-ranging and international group, including people with lived experience. The expert group comprised practitioners and researchers with backgrounds in PS, psychiatry, nursing, psychotherapy, and clinical psychology from Australia, Canada, England, Germany, Japan, Scotland, and the USA. Their expertise spanned PS trials, intervention development, and working as PSWs. In January 2023, experts (n = 21) were sent an initial list of publications and asked to identify further publications to be considered for inclusion.

Once the included document list was finalised, the data abstraction table (DAT) was completed for all included publications. Data extracted comprised authors, year, type of publication (e.g. trial report), country, service setting (e.g. community or inpatient), population, and candidate PS components (name and description).

The identified candidate PS components were analysed by the core analyst team (YK, AC, CN, FN, MS), using tabulation to de-duplicate, integrate, and thematically group the list, in order to create a preliminary typology comprising an un-ordered list of components and sub-components. All the components were reviewed and were grouped into initial themes that presented common ideas. Each initial theme was defined, then was refined for coherence and distinctiveness (e.g. the wording of each theme, the place in the typology, the definition of each theme) (Kotera et al., 2021). The preliminary typology was then refined through consultation with 21 experts from seven countries, chosen for their expertise in PS practice and research. In March to April 2023, the experts were asked to review the DAT and offer comments on the structure, coherence, and language, and to identify any important missing components or sub-components. The typology content and structure were refined based on this expert consultation.

Results

Overall, 42 publications were included, comprising 26 trial reports, nine training manuals, and seven change model papers.

Trial Reports

First, nineteen articles in White’s systematic review (White et al., 2020) were all included, and 532 articles were retrieved through database searching (Medline 222; PsycInfo 310). One hundred and one articles were removed as duplicate. The remaining 431 articles were title/abstract screened, which identified 47 articles to be full-text reviewed. Six articles were identified as includable trial reports. Main reasons for the excluded 41 articles were (a) not being an RCT, (b) not PS, (c) participants not being in a mental health service, (d) not one to one, (e) not measuring a mental health outcome, and (f) reporting from the same trial already listed. Through expert consultation, one more article was added.

PS Training Manuals

Fourteen documents were initially identified from online searches. Eight articles were removed as they were not training manuals. Experts identified three additional training manuals. In total, nine training manuals were included.

Published Change Models

Four change model papers were initially identified. Three papers were added through expert consultation.

All included publications are numbered and listed in the Electronic Supplementary Material 2.

Typology

The DAT was created to characterise the 42 included publications and to record the 242 long list of candidate components identified from included publications (Electronic Supplementary Material 3).

The core analyst team thematically categorised the identified candidate components and created an initial typology. The initial typology comprised 16 components and 11 sub-components, which were categorised into four themes: recruitment, preparation, practice, and PSW wellbeing. After expert consultation, three sub-components (“activity log”, “workbook”, and “fidelity check”) were removed, one component was transferred to another theme (“service user match” moved from preparation to practice), and one component changed its name (“recording” to “record-keeping”). No change was made to the four themes. The final typology comprised four themes, containing a total of 16 components and eight sub-components. Table 1 presents the final typology with definitions and example texts for each component and sub-component.

Table 1 Typology of one-to-one mental health peer support work components
Full size table

Discussion

This systematised review and expert consultation developed a typology of one-to-one PS components through reviewing trial reports, training manuals, and change model papers. A long list of 242 candidate components were identified from 42 publications, which were then integrated into 16 components and eight sub-components, organised into four themes.

The novelty of this review was developing a typology of PS components, validated through expert consultation. To date, there is no standardised approach for assessing which part of PS has impacted service user mental health, and to what degree. This reduces the opportunity to compare findings across different PS studies. Being unable to compare the findings across studies is concerning, because PS is actively used in different contexts and regions of the world (Borthwick et al., 2020; Pearson et al., 2021). This typology allows assessment and comparison, which can help support and advance PS in both practice and research.

In practice, this typology can help reflection and planning. PSWs can use the typology to reflect on their experiences, which inform adjustments in their practice. The importance of self-reflection is increasingly recognised among people who offer care for others (Kotera et al., 2022). Each PSW can reflect on their experiences in each component. This can facilitate more focused self-reflection, peer discussion, and supervision. Peer discussion and supervision are often used in PS and are an essential part of successful PS (O'Connell et al., 2018; Rogers et al., 2016).

This typology can be used to plan PS clinically and financially. Clinically, the typology can inform session plans. For example, service user information may not be available in some contexts, which can hinder the service user match (component 3.1). To compensate this hindrance, PSWs can plan to focus on other components such as the relationship building (component 3.2) to maintain the impact on their service users. This is especially helpful to PSWs. Financially, the typology can help PS workforce planning. Service managers can identify the key infrastructure elements that need to be considered in developing a PSW workforce such as recruitment processes, training, and supervision arrangements. The typology can help assess what needs to be costed in order to implement PS.

In research, this typology allows more rigorous and synthesised analyses of PS including CNMA. CNMA evaluates the impact of each component in a complex intervention, answering questions such as “Which components work (or do not work)?” (Sofia et al., 2022). CNMA also assesses the intervention without some components by disengaging them from a model (Sofia et al., 2022). Such rigorous and synthesised analyses can advance PS research because inconsistent reporting is one major problem of PS research (Richard et al., 2022). The results can inform the most effective components or combination of components. This may be useful to many PS organisations, especially to those that are under-resourced as they are informed of which components to focus their limited resource on. The findings can also help identify specific needs for training and supervision.

The typology also can reduce a gap between practice and research in PS. Poorly labelled and/or described interventions are a barrier for implementation, because readers, who may be healthcare workers, service users, and others, cannot be certain of what they mean and what to do (Hoffmann et al., 2014). The importance of comprehensive labels and descriptions is an important yet under-researched area in implementation science (Hoffmann & Walker, 2015). Our typology offers comprehensive language to PS, facilitating its implementation. This can bridge PS practice and PS research.

Further evaluation is needed when using the typology in different contexts. The typology is based on RCTs of one-to-one PS for adult mental health service users. PS is used in many different contexts (e.g. forensic contexts), targeting people with other characteristics (e.g. elderly, youth, LGBTQ+, disabilities) (Borthwick et al., 2020; Shalaby & Agyapong, 2020; Williamson & Durcan, 2020). The utility of our typology in these applications remains to be known. This highlights a need for more rigorous syntheses to identify effective components and harmful components, to inform approaches for better applications into other contexts and populations.

Strengths and Limitations

The strengths of this review include the expert consultation and the use of three different sources of information. Twenty-one PS experts from seven countries were involved to ensure our findings retain a comprehensive, balanced perspective reflecting different cultural imperatives. This strength was further enhanced by referring to three different information sources, spanning the various contexts in which change models can be reported. Several limitations should be noted. First, the searches were conducted in the English language only. Second, many of the experts were from a western and high-income country. Third, as this was a systematised review, some relevant documents might not have been included. This includes resources from other established organisations that offer PS such as the Substance Abuse and Mental Health Services Administration and the National Alliance on Mental Illness. However, we judged additional papers to our detailed analyses, involving expert consultation and identification of candidate components, would not change the output. Fourth, the details of some components still need to be assessed. This includes what to match in the service user match (e.g. age, gender, and life experiences), and differences between supervision by a peer and by a non-peer, as non-peer supervision is also found helpful to PSWs (Forbes et al., 2022). Despite these limitations, our systematised review and expert consultation are the first rigorous, comprehensive work, offering helpful and timely insights into components of one-to-one PS.

Conclusion

This systematised review and expert consultation identified 16 components and eight sub-components in one-to-one PS. The typology can help improve the practice and research of PS by enabling practitioners to facilitate their reflection and planning, and researchers to conduct more synthesised analyses of PS. The typology can also reduce the practice-research gap by offering a comprehensive and coherent language. Studies such as CNMA need to be conducted to further understand how the various active elements of PS interact to contribute to beneficial outcome.