This study was based on the definition of SV developed in the healthcare sector (4). By extension, this conceptual framework was applied to other high-risk professions, defining an equivalent concept as a highly stressed worker facing a situation that exceeds their resilience and coping abilities.

Professions such as emergency responders, pilots and train drivers frequently encounter high-pressure situations where errors can have serious consequences, leading to emotional distress and trauma. While healthcare professionals typically face unanticipated adverse patient events, unintentional healthcare errors or patient injuries, operators of hazardous machinery, pilots and train drivers encounter unanticipated operational incidents, critical system failures, unintended safety breaches or high-risk operational errors. These events may arise from unexpected technical malfunctions, unintended procedural deviations or adverse operational events, requiring professionals to respond swiftly to unforeseen human or mechanical errors, safety-critical incidents or high-stakes decision errors. Despite the differences in context, the psychological and emotional impact of such events often follows similar patterns across high-risk professions.

First, we conducted a preliminary search to identify existing systematic or scoping reviews and to refine our inclusion and exclusion criteria and keywords. Databases searched included PubMed/MEDLINE, the Cochrane Database of Systematic Reviews and JBI Evidence Synthesis. Next, we redefined the inclusion and exclusion criteria using the population, context, concept (PCC) framework.30 Finally, we conducted a comprehensive search in February 2024, covering the following databases: PubMed/MEDLINE, ProQuest, Scopus, Web of Science, PsycINFO and EMBASE. Of these, PubMed/MEDLINE, Cochrane and JBI Evidence Synthesis are health-specific, while ProQuest, Scopus, Web of Science, PsycINFO and EMBASE are generalist databases that include research from multiple disciplines, such as social sciences, technical fields and industrial sectors. This diversity of databases, combined with the use of field-specific terminology, allowed for the exploration of studies across multiple disciplines, including technical, social and healthcare fields. This approach facilitated the identification of relevant studies to effectively address the research question of this study. The search strategy used is found in table 1.

The structured search was carried out by two researchers (CP-E and AS-G) in parallel, who then met to share their results and reach a consensus on the terms to include in the search strategy. The search strategies were refined through team discussions and exported into EndNote, where duplicates were removed. In some databases (ProQuest, Scopus, EMBASE), we used specific filters to improve the search specificity (table 1).

We also reviewed articles from the bibliographies of eligible studies and institutional websites, including those of the European Researchers’ Network Working on Second Victims (ERNST), AHRQ, WHO, Segundas y Terceras Víctimas Proyecto de Investigación (Research Project on Second and Third Victims), SARS-CoV-2 (COVID-19) Second Victims, Centre for Patient Safety, Second Victim Support (UK) website, KU Leuven Research–Second Victim in Healthcare, ForYOU team website, AHRQ website, AHRQ PSNet and WHO website.

Two of the included databases (ie, ProQuest and Scopus) already included grey literature. For this reason, no additional search for this type of reports was conducted.

The selection process consisted of three stages carried out in parallel by two reviewers (CP-E and AS-G):

A consensus was reached to resolve conflicts found between the reviewers. This phase was carried out using Rayyan software, a tool used to collect and examine bibliographic references.

The entire data extraction process was again performed by two reviewers in parallel (CP-E and AS-G), who carried out standardised steps to analyse and synthesise the data from the selected sample of articles. Any discrepancies were then analysed and resolved through consensus meetings to ensure reliability.

Reviewers extracted data encompassing various details, including authors, publication year, country, document type (eg, original research article, review), study classification (experimental, observational, etc), stated study objectives, industry context (eg, healthcare, public safety, etc), participants (professional profile, age, or other relevant characteristics, along with sample size) and identified competencies (skills, attitudes or knowledge). Moreover, we captured information regarding the specific programme or intervention under study (eg, RISE, FOR YOU) and the study outcomes, with a specific focus on results related to the assessment of supporter competencies (data extraction template can be consulted in online supplemental file 1).

As a measure of the impact of each publication included in this study, we considered both the Journal Impact Factor (JIF) and JIF Rank of the journal in which the study was published, as well as the number of citations received. Thus, the data in online supplemental file 2 provides insight into the study’s significance within its relevant scientific domain.

Once we selected the resources, we extracted and listed the competencies from each resource. Then, we categorised them, grouping them by domains or types of competencies, and counted the frequency of each one. In order to ensure clarity in the categorisation of competencies, we followed a systematic approach that involved an iterative review process among multiple researchers (JJM, CP-E and AS-G). Each competency was assessed in the context of its primary function, with particular attention given to domains that appeared to overlap. When competencies showed clear links to two or more categories, a decision was made based on their core focus. This iterative process aimed to minimise ambiguity and ensure that no competency was misplaced. The final categorisation, including domain definitions, is presented in online supplemental file 3. Additionally, we compared these competencies across the professional sector.

A total of 1237 studies were initially identified through the database searches. Following full-text screening, 20 studies were ultimately included. In addition, a citation review identified 55 further studies for evaluation, of which 14 were finally included. In total, 34 articles were included in our review. A PRISMA-ScR flow diagram visually summarises the study selection process (figure 1).

The included articles were published from 2005 to 2024, with most conducted in the USA (n=18, 52.9%).31–48 The rest of the studies were conducted in the UK (n=4, 11.8%),49–52 Canada (n=4, 11.8%),53–56 Australia (n=3, 8.8%),57–59 Germany (n=3, 8.8%),60–62 and one each from New Zealand63 and Slovenia (n=1, 2.9%).64 Most of the articles adopted an observational study design, frequently using questionnaires, observations or interviews (n=27, 79.4%). There were also experimental studies (n=3, 8.8%), quasi-experimental studies (n=3, 8.8%) and a narrative review (n=1, 2.9%). The studies covered a range of industrial sectors, with the majority focusing on healthcare (n=20, 58.8%), followed by public safety personnel, including police, firefighters, emergency personnel and similar roles (n=8, 23.5%); the military sector, including active professionals and veterans (n=5, 14.7%); and postal services (n=1, 2.9%). A summary of the included articles is provided in online supplemental file 4.

A total of 29 peer support programmes were found, including seven unspecified or ad hoc interventions, as detailed in online supplemental file 5. Many programmes aimed at providing support to professionals facing stressful and traumatic situations. Additionally, there were specific programmes designed for adverse events in healthcare and post-traumatic stress disorder in veterans, and initiatives focused on increasing resilience and well-being among professionals at high risk of exposure to potentially traumatic events (online supplemental file 5). Only 13 studies (38.2%) mentioned the term SV, all of which were in the healthcare context. None of the studies from other industries used this term, and seven studies based on the healthcare industry did not explicitly mention it.

A total of 302 peer supporter competencies were identified. Through the categorisation process, these competencies were grouped into 30 general, broader competencies, such as communication skills. The general competencies identified were included in five domains: communication and interpersonal competencies (competencies related to the skills for effectively interacting with peers while providing support), emotional support and intervention competencies (competencies related to providing emotional support and first aid to deal with negative emotions), ethical standards and legal support competencies (necessary ethical standards for the role and knowledge to offer legal support), organisation and role related competencies (competencies related to the role as a peer supporter and understanding of the organisation) and self-care and motivation competencies (competencies related to the ability to care for oneself and to provide care and encouragement to others). These domains were defined by the researchers based on an iterative analysis of the extracted competencies, grouping those that shared characteristics or referred to similar themes. These broader competencies are summarised in table 2. A detailed description of each domain and their corresponding competencies is provided in online supplemental file 3.

Ninety-one unique, specific competencies were found once similar, and overlapping competencies were combined. Online supplemental file 6 provides a detailed overview of the general competencies, their specific subcategories and details regarding their frequency across articles and sectors.

The reviewed articles also revealed a series of requirements for peer supporters that could not be considered competencies. Among these, 24/7 availability to respond to colleagues’ needs was found to be particularly relevant for those assuming the role of peer supporter,36 47 48 52 63 indicating the high level of commitment required. The aspect of being a volunteer was also predominant.47 49 51 Other requirements included not suffering from mental health issues,49 51 having experienced a similar trauma,39 being at least 30 years old,64 or having a minimum of 5 years of work experience.64

The instrumental, attitudinal and knowledge-based competencies that peer supporters should possess are directly related to the psychological first aid techniques they are expected to perform. Specifically, they should be capable of practising active listening, reframing and reflective listening and be able to communicate clearly.31–38 40–45 48 49 51 53–57 59–64 Peer supporters must exhibit empathy, demonstrated through openness31–33 35 38 41 42 44 49 51 54 64 and be familiar with the resources available within the institution to support the SV.32–35 38 40–48 50 52 61–63 They should also be knowledgeable about strategies for managing stressful situations33–35 41 42 50 53 55 56 58–62 and promoting self-care.32 40 42 48 56 57 63 Lastly, they must have a natural leadership ability that inspires trust in others.39 41 44 47 49 51 52 54 57 63 64 This profile has been found to be consistently demanded across various industries considered in this study, including healthcare, armed forces, firefighting and police.38 41 42 51 55 64

Interestingly, although this review did not limit the type of industry, the 34 studies included cover only eight different industries: healthcare, military, postal services and public safety personnel, further divided into profiles such as police, firefighters, rescue and protection, correctional officers and a mixed category where these and other emergency profiles are combined. This review missed representation from sectors such as aviation, transport and energy, from which no work was extracted. In both scientific and grey literature, different sectors have highlighted the importance of maintaining a zero-error environment, identifying the emotional impact following traumatic events or errors similar to those known for SVs and developing interventions to help them recover.65 This may be due to the difficulty of identifying SVs without a common term across all industries.65 It may also be because, in these sectors, while interventions to reduce the distress of SVs are being implemented, peer support programmes are not yet well developed, and the competencies required by peer supporters have not been studied. Similarly, while the healthcare sector has produced a substantial amount of literature on peer support programmes for SVs,3 66–69 there is limited coverage on the specific competencies that need to be trained by peer supporters.

Despite the different industrial sectors reviewed, there is significant consistency in the competencies identified. This is partly because the response to highly stressful situations is a human condition shared regardless of the sector, although the intensity may vary depending on the perceived severity of the threat. Interpersonal communication skills were consistently identified as essential competencies across all studies, indicating that these skills are necessary for peer supporters regardless of the industrial context or profession. Additionally, psychological assessment and triage were frequently pointed out, suggesting that peer supporters need to perform a basic evaluation to assess the severity of SVs’ symptoms and refer them to higher levels of care if necessary. Another frequently identified competency is the knowledge of stress and trauma, implying that peer supporters should have a basic understanding of the symptoms and signs of stress and the functioning of trauma. Moreover, peer supporters must maintain a non-judgmental attitude and confidentiality, which is critical for professionals to seek help without feeling judged or fearing for their reputation. Empathy and credibility as professionals were found to be equally important across different contexts.

However, essential differences were found in the industrial sectors analysed, shaped by each industry’s specific context. For instance, in the military and healthcare sectors, a deep understanding of organisational structure is crucial,41 59 whereas in postal services, other specific aspects are emphasised.53 In sectors like firefighting and veterans’ affairs, training includes addressing substance abuse and alcohol issues, with a particular focus on suicide prevention.38 41 Interestingly, only the healthcare sector incorporates practices such as self-care,32 40 42 48 56 57 63 resilience building,42 56 57 60 61 mindfulness56 63 and the importance of non-investigative approaches to healthcare errors.32 33

Peer supporters require basic training to effectively support SVs who place their trust in them. This study highlights that having experienced being a SV is not sufficient. Training in active listening, reframing and reflective listening techniques ensures that peer supporters can communicate effectively and be understood. According to the 5-tier ERNST model,70 training should also include acquiring competencies to share stress management and self-care strategies, as the SV needs to reinforce resilience and enhance self-care strategies after the critical phase. Highly stressful situations are recurrent in professional careers.71 72 A resilient professional is better equipped to cope with stressful events, showing greater emotional balance and lower psychological burden.73 74 Moreover, structured peer support interventions have been associated with improved emotional recovery and enhanced professional resilience.10 75

This training should combine various active learning techniques such as case studies, simulations and role-playing scenarios. Based on these findings, training should be accompanied by acquiring knowledge about available resources to provide proper guidance, including knowing when to refer to mental health services. A possible structure would allocate at least 40% of the content to practical training, delivered through simulations, role-play exercises and supervised practice. Participant selection criteria should ensure an appropriate professional background as well as a demonstrated empathetic profile. The training programme could incorporate standardised assessment tools, ongoing supervision and external certification, aligning with emerging evidence and established best practices. In terms of content, it would be advisable to integrate elements of Psychological First Aid, especially for the initial response to distress, as well as to consider the ERNST model,70 which includes a strong emphasis on prevention and system-level responses. These frameworks can guide the development of a comprehensive and structured training model.

Customised training content should address specific scenarios and challenges relevant to healthcare, industry, armed forces, firefighting and police. Building a team of trainers from diverse industries could enhance these programmes and the capacity of peer supporters, as they face similar human condition-related problems across these fields.

Implementing a system for ongoing evaluation and feedback would allow for necessary adjustments based on peer supporter performance and feedback. This aspect should be further explored in future studies to deepen the understanding and effectiveness of peer support training programmes.

Our review focused on mapping the core competencies necessary for peer supporters to assist SVs. Given that the term SV has been predominantly used in the healthcare sector, we did not limit our search to this term alone but extended it to its meaning. To address this limitation, we employed various databases from different disciplines, ensuring that the terms used were not restricted to the healthcare sector. This approach aimed to include more studies on the competencies of individuals who provide initial support to colleagues feeling responsible for having caused or being about to cause significant harm to others (eg, following a medical error, an inappropriate manoeuvre, insufficient maintenance). However, the challenge of identifying this type of individual without a universally accepted term may have still constrained our ability to fully capture all relevant aspects. Additionally, the initial list of competencies was not subjected to a critical secondary review by a group of experts, which could have provided a more rigorous validation of the results. The subjectivity of our process, despite our efforts to be systematic, may represent a potential limitation in the reliability and comprehensiveness of the study. Furthermore, although it is not mandatory for a scoping review, we did not carry out a critical appraisal of the quality of the sources consulted. We believe that future research should include a critical assessment of the quality of the sources and validation of the identified competencies and a consultation with experts in the field.

To assess the quality of the included studies, we chose to analyse indicators of journal quality and the publication’s impact within the scientific community, such as the impact factor, the JIF Rank and the number of citations each study had received at the time of analysis. Future research on this topic could adopt a standardised evaluation of evidence strength, using established scales or methodological frameworks to ensure a more systematic and rigorous assessment.

Not applicable.

Not applicable.