Lived experience peer support programs for suicide prevention: a systematic scoping review

A research team was convened consisting of three researchers in field of suicide prevention (MS, AM, LR) and one researcher with lived experience and consultancy roles in mental health and suicide related peer-support (IO). Using a co-design methodology, the team met to discuss the purpose of the review and was guided by IO’s experience in the peer support sector of suicide prevention in developing the review protocol and research questions [26, 27].

Two exploratory research questions were developed: What types of suicide prevention peer-support programs delivered by people with lived experience currently exist? What do we know about their effectiveness? These broad questions allow us to generate an overview of research undertaken on this topic and to identify where the research gaps in peer-led support programs in suicide prevention lie.

We undertook a systematic search for peer reviewed articles, a search of grey literature databases, and a website search and expert consultations to identify eligible programs. A systematic search of the literature was conducted for articles published between 2000 and 2019 using Medline (PubMed), PsycINFO (OVID interface), and EMBASE (OVID interface). Bibliographies of previous systematic reviews and included papers were also searched. Grey literature was searched to include research that had not been peer-reviewed, including Google Scholar and the World Health Organization (WHO) Clinical Trials Registry (limited to a 5-year period from 2013 to 2018). To identify existing suicide prevention peer-led support programs, we also approached clinical and academic content experts and searched relevant internet sites including organisations known to be active in suicide prevention. See Additional file 1: Table S1 for a list of identified and screened websites and programs.

Search terms were developed relating to the three key concepts underpinning the literature review: suicide, peer support and lived experience. These were in alignment with our definition of lived experience peer-led support programs in suicide prevention. Alternative terms for peer support were developed and refined during iterative test searches. The final search strategy was developed using medical subject headings and free text words related to peer support and suicide prevention. Search terms were adjusted to fit the requirements of different databases. A complete list of search terms by database is available in Additional file 2: Table S2.

All records were imported into Endnote (version X8.2) and screened for inclusion in two stages. In the first stage, one researcher (AM) screened titles and abstracts for potential inclusion. In stage 2, the full texts of retrieved articles were screened independently by two researchers (AM and MS). Discrepancies were resolved in a meeting between the two researchers. Google Scholar records were screened first by title and abstract and then by full text by MS. WHO Clinical Trials Registry entries were screened by LR. Grey literature was retrieved by IO via internet search and consultations with experts in the area and then screened for inclusion in a team meeting by IO, LR, MS and AM.

Records were eligible if they included a peer-led support program with a focus on suicide prevention for people who experience suicidality. Peer supporters had to have lived experience of suicide; they could be community volunteers, people with similar experiences, or health professionals/health care staff if they had a lived experience of suicide and this informed their support role as peer supporter. Our review was not restricted to programs with matched lived experiences. There were no restrictions on the delivery mode of the programs; for example, programs could be delivered one-on-one, in group settings, as telephone support, online, at home, or in respite care.

We excluded records if programs were not delivered by people with a lived experience of suicide or where this could not be determined from the program description; were suicide bereavement programs; were capacity building or workforce training programs such as gatekeeper training, suicide awareness raising or suicide literacy programs; were focussed on improving mental health more broadly; or were a component of a multi-component intervention and not described separately within the larger program.

Articles in academic databases and Google Scholar had to be published between January 2000 and August 2019 and trials had to be registered on the WHO Clinical Trials Registry between 2013 and 2018. Academic databases were first searched on 14 June 2018. A second search of the academic databases was run on 29 August 2019 after the authors became aware of new evidence published since the original search. Google Scholar was searched on 21 June 2018 and the Trial Registry was searched on 7 September 2018. There was no time restriction on programs and records identified through expert consultation and web searches. We included any evaluation reports of eligible programs, irrespective of study design, setting, participant age, and publication language, so long as they could be translated into English. The Flow Diagram in Fig. 1 includes the number of records at each screening stage for all data searches and data sources combined.

Following these criteria, we identified 4077 records through electronic searches and 408 through searching other sources including grey literature, websites and expert consultations. Following removal of duplicates, 3058 records were screened by title and abstract information. This led to the exclusion of 2925 records. The remaining 133 records were read in full text by AM and MS who reached consensus on the final inclusion of 8 records according to the selection criteria. Of these 8 records, two referred to the same program, and were therefore summarised as one program in the results section.

Records identified for final inclusion were extracted into a charting table which documented the following information: Reference, title and country of the program; program description including setting modality and lived experience mode; study aim and methodology; sample characteristics and key findings (if the program had been evaluated). Table 1 lists details on all identified programs.

The information in the charting table was then synthesised in accordance with the two research questions of this paper. Due to the variation in study design and the absence of evaluation data for many identified programs we focussed on a narrative summary of studies. First, we provide an overview of the types of peer support suicide prevention programs (research question 1) by briefly summarising programs by their setting, modality and the role of lived experience. Secondly, for those programs that had been evaluated we characterise the study aim and methodology and discuss key findings regarding their effectiveness for suicide prevention (research question 2) (Table 1).

Our search identified eight records that fulfilled the inclusion criteria. These described seven programs that were from four different countries: USA [3], Germany [2], China [1], and Australia [1], and which varied in setting and design (see Table 1). Of the seven programs, four included an evaluation component [28‐32] and three provided descriptive accounts of a program [33‐35]. Four peer support programs provided group support [28‐31, 34], two were designed to deliver one-on-one support [32, 33] and one program included mixed modes of support [35]. Regarding the program settings, two programs were delivered in clinical settings [32, 35], two in the community [28, 33], two online [29‐31] and one in schools [34].

Salvatore [35] presented a mixed mode peer support program within a psychiatric hospital. It was offered to patients of the hospital and their families. The hospital employed two peer support staff to deliver one-on-one peer support as well as group support. While this program has been implemented as part of the Montgomery County Emergency Service, evaluation data on its effectiveness has yet to be published. The second program designed for patients in clinical settings was the Peers for Valued Life (PREVAIL) program [32]. This program is a one-on-one support service for people who had attempted suicide and were patients of a psychiatric ward. Patients were teamed up with a peer specialist with lived experience of suicide and weekly meetings were held for up to 12 weeks after discharge from the ward. Peer specialists received training on risk assessment, using suicide prevention tools, communication and relationship building and motivational interviewing. In situations of acute risk clinicians were contacted. The aim of this program was to reduce suicide risk post-discharge from a psychiatric ward.

The two community-based suicide prevention programs were designed to provide support to those in crisis or experiencing suicidal ideation. Firstly, the Alternatives to Suicide program (USA) runs peer support groups for people experiencing suicidality [28]. Open group discussions are facilitated by trained peer supporters to enable conversations around the reasons and factors that may have contributed to someone wanting to die. Reflective of the key principles of Validation, Curiosity, Vulnerability and Community the conversations are non-judgemental and free of boundaries. Groups provide a safe and comfortable space to talk and focus on offering a non-clinical environment to build trust. Secondly, the Peer CARE Companion Program (Australia) offered through the Way Back Support Service is a new program which is currently being trialled [33, 36, 37]. It is directed at supporting people with a lived experience of suicide (experiencing a suicidal crisis or after a suicide attempt) in one-on-one peer support settings. The program was developed through a collaboration between three mental health organisations in Australia (Beyond Blue, New Horizons and Roses in the Ocean). Two trials and a consultation process including people with lived experience in 2017 and 2018 led to a revision of the program and the results of a second trial are yet to be released.

Two programs were set online using data from online messaging boards [29‐31]. Both aimed at better understanding the benefits and risks of participation to people experiencing suicidality by looking at the effect that using online messaging boards has on participants. These online messaging boards are best described as informal group support interventions. While meaning to support people who experience suicidality, due to their open entry format they allow both people with lived experience and non-suicidal people to participate. The content is participant/online user driven with limited control for quality and safety for people at risk.

Finally, a school-based program aimed at early detection of at-risk youth in Chinese schools, offered peer group support sessions led by students and supported by teachers and the school community [34]. Students at risk were identified by peers or teachers and invited to participate in student-led support groups that met regularly. Teachers visited the groups monthly to help address any issues if needed. Group membership was voluntary and group leadership rotated. Group members were taught how to recognise unhappy and depressive behaviour in peers. Group leaders reported to the teacher and were able to refer students further if concerned.

With the exception of the school-based peer support program, which broadened its scope to include mood, depression and self-esteem as early warning indicators of suicidality, all programs were specifically aimed at supporting people with a lived experience of suicide. It was however decided to include the school based intervention as its overall goal was described as reducing suicide risk in youth.

Three of the seven identified programs contained evaluation data; two were quantitative studies [28‐30], one was qualitative [31] (see Table 1).

The Alternatives to Suicide program reported early findings from internal feedback surveys with attendees of the support groups [28]. Findings indicate that attending the groups was perceived as helpful as participants felt that they could talk freely about their experiences. Attending the groups had improved at least one area in attendees’ lives. Areas of greatest improvement were increased sense of community and a better understanding of why suicidal thoughts may come up.

Kral and Eichenberg found that participation in online peer support forums decreased the intensity of suicidal thoughts [29, 30]. The authors collected data from participants of the online forums via an online survey. Thirty-one percent of participants self-reported a decrease in intensity of suicidal thoughts as a result of their participation in the messaging boards. While 22% of respondents said they were more motivated to seek professional help, using suicide messaging boards did not increase help-seeking outside the forums. The main motivation for using online forums was for emotional support, to feel understood and receive comforting reactions.

In a qualitative analysis of threads from a suicide online forum Niederkrotenthaler and colleagues [31] found that participation in this forum can help to improve a person’s suicidality. The authors downloaded threads from seven pre-identified suicide message boards and thematically analysed a random selection of these threads. Several communicative strategies were associated with psychological improvements in online forum participants; these were receiving constructive advice, being actively listened to, receiving sympathy for one’s struggle, and provision of alternatives to suicide by other members of the forum.

While the Pilot Randomised Controlled Trial of the PREVAIL program did assess the programs feasibility and acceptability by collecting quantitative and qualitative data on peer specialist performance, this trial was unable to assess the programs efficacy in reducing suicidality in participants due lack of power [32].

In screening the literature, we identified that 84 records had to be excluded in the full-text review stage due to either not providing enough information on the nature of the lived experience of peer supporter workers or because lived experience was defined more broadly and not specific to suicidality. For example, it was frequently unclear if a peer support program was in fact peer-led or led by a clinician, and when the program was peer-led it did not specify if the peer supporter had lived experience of suicide or whether this was defined more broadly in the context of mental health. It is possible that the lack of clarity in the definition of peer support may have led to the exclusion of otherwise eligible programs during the screening process of this review.

Further, we found that in some programs the definition of peers was not aligned with our selection criteria. Some programs described their intervention as peer group support, yet the group was led by a health professional [40]. Others had a peer supporter as co-facilitator alongside a leading health professional, therefore not qualifying as a peer-led program [41]. In particular, school-based programs working with students tended to select peers on the basis of age or belonging to the same social group but did not make suicidal experiences part of the condition for becoming a peer supporter [42, 43].

Despite the effort that has gone into defining peer support and lived experience in the context of suicide prevention in recent years [2, 10], this seemingly has not yet translated into research designs and publications. The findings from this review highlight that authors follow varying definitions for peer-led support and often fail to provide adequate detail in the description of their program about what constitutes peer-led support in their respective program. This limits our understanding of the nature of peer support within existing programs and ultimately affects what implications we can draw from existing literature on the effectiveness of peer-led support suicide prevention programs.

To advance knowledge on peer-led support programs in suicide prevention we suggest a few areas for future investment. Firstly, the development of a framework for standard reporting on peer support initiatives would greatly improve our understanding of the breadth and depth of current peer support programs [44]. In addition, improved quality of reporting on peer support roles in suicide prevention programs would help to clarify the underlying definition of peer support. Secondly, we need high-quality evaluations of peer support suicide prevention programs and of peer-led components within larger programs to better understand their effectiveness on participant health across different settings and delivery modes and to allow for comprehensive systematic reviews and meta-analysis in the future. This evidence can help enhance our efforts to better integrate peer-led support with conventional crisis support and find mutual benefit in both. Thirdly, while peer support is generally accepted as a positive addition to care by legislative bodies, we currently lack models for the efficient and effective integration of these programs alongside conventional care [38, 45]. Addressing this issue would facilitate peer support to become a care component in its own merit.

It should also be noted what is already known about the positive effects of peer-led support in other related areas. While the knowledge is scarce on peer-led suicide prevention programs, it could be beneficial to revisit evidence from mental health peer support and investigate whether similar approaches could be adapted to suicide prevention. This practical approach could then be subject to further testing and refining to cater to specific needs in suicide prevention.