Peer support groups after acquired brain injury: a systematic review

ABSTRACT Background Peer support groups can be a way to obtain support, problem solve, and widen social networks. However, there has been no systematic literature review examining the evidence for the use of peer support groups after an acquired brain injury (ABI). Objective This review sought to systematically evaluate the evidence for (1) the psychosocial effectiveness, and (2) the experience of peer support groups in adults who had experienced ABI’s. Methods The systematic literature search was conducted across the following four databases: PsycINFO, PsycARTICLES, MEDLINE, and Cumulative Index to Nursing and Allied Health Literature (CINAHL) in October 2019. The mixed methods appraisal tool was used to examine the quality of the research. Results 13 papers were included in this review. Limited evidence was found for the psychosocial effectiveness of peer support groups in ABI, although the experience of partaking in a peer support group was largely found to be positive. The benefits and helping factors of taking part in a peer support group can be summarized as: being connected, interacting with others, and providing and receiving support. Conclusions The findings of this review suggest that peer support groups could be a promising intervention to support individuals and promote adjustment following an ABI.


Introduction
As an adult, adjusting to life following an ABI can involve managing: activity restrictions (e.g. driving or employment), increased reliance on family, changes within roles and relationships, and financial hardship (1,2). Many of these changes have been shown to have a negative impact on sense of self, as individuals progress through their recovery and realize they may not be able to return to work, hobbies, and other aspects of their pre-injury day to day life (3). Whilst services such as neurorehabilitation units, community teams, and voluntary organizations exist to help support the process of rehabilitation and adjustment, another potentially useful and relevant intervention following an ABI is peer support groups.
In its simplest form, peer support can be defined as being composed of individuals who share a similar problem and come together to provide mutual help and support (4). A review by Dennis (5) found that the way in which peer support is offered varies considerably across a range of factors including: intervention type (e.g. one to one mentoring, self-help groups, support groups, and online groups), provider (e.g. community or hospital based, and voluntary organisations), and structure (e.g. highly structured or informal interventions).
Reviews outside of an ABI population have documented the positive and unique impact that peer support can have on individuals. Davidson (6) reported that peer support offers three contributions: (1) the installation of hope using selfdisclose, (2) a modelling function of self-care behavior, and (3) a relationship between peer provider and recipient which is characterized by trust, acceptance, understanding, and the use of empathy. Further research, specifically focusing on peer support groups, has cited positive outcomes including: a unique sense of community, the opportunity to be accepted, and the opportunity to gain information from others (7), as well as to socialize and broaden social networks (8).
Based on the above research and the well-documented negative changes that individuals post-ABI can experience, peer support seems to be an appropriate intervention to offer to adults post-ABI. The flexibility and variety in the way peer support is offered makes it a viable intervention which can be easily tailored to the resources of individual services, and the positive outcomes, especially the opportunities to share and problem solves have been documented as desired by individuals post-ABI (9).
In terms of previous reviews which focused specifically on peer support within ABI, two papers were found to be relevant. The first was a review by Wobma et al. (10), which examined the evidence for peer support in rehabilitation following ABI. This review only included randomized control trials and as such only included two papers, meaning that the findings were limited. An additional review was conducted by Morris et al. (11), which focused on the use of peer mentoring in traumatic brain injury. This review highlighted six papers, and reported some positive effects on behavioral control, mood, coping, and quality of life, but with limited evidence on the whole. Therefore, whilst peer support has been documented to have a range of positive and unique consequences in other populations, previous reviews have found limited evidence for the use of peer support following ABI's. This appears to be a reflection on the small number of papers included in each review, rather than evidence against the use of peer support following ABI. Given that a previous review has already been conducted into peer mentoring, this review seeks to evaluate the evidence for the use of peer support groups after ABI, considering both quantitative and qualitative research. For the purpose of this review, peer support is defined as 'being composed of individuals who share a similar problem and come together to provide mutual help and support' (3), and a group is defined as more than two people.

Eligibility criteria
Studies were included if they met the following criteria: (1) employed a peer support group for individuals with ABI; (2) the sample was obtained from individuals who had experienced an ABI (ABI occurred >18 years). If the sample included family members or carers, the data were included if more than 50% of the data was from individuals who had experienced an ABI. This was chosen at the discretion of the reviewer as it was observed that several of the groups included reported family members/ carers as a participant, and excluding on the basis would limit the number of papers included. The 50% marker was chosen to ensure that the majority of the data were obtained from participants who had brain injuries; (3) was published in the English language to eliminate errors with translation; (4) the study was published in a peer-reviewed journal article to try and ensure the papers included were of good quality; and (5) published any time up to the 28 October 2019.
Papers were excluded if: (1) they used a one-to-one model of peer mentorship, rather than a peer support group, as this has been covered in a previous literature review; and (2) papers that reported data about self-management groups or group therapy (e.g. a CBT group) for individuals who have experienced an ABI, as these have distinct aims that differ from peer support groups.

Information sources
The systematic literature search was conducted across the following four databases: PsycINFO, PsycARTICLES, MEDLINE, and Cumulative Index to Nursing and Allied Health Literature (CINAHL).

Search
During a scoping search, the search terms of relevant papers were examined to help identify the search terms most likely to identify articles relevant to this review. The following search terms were used to search article titles and abstracts: ("brain injur*" OR "head injur*"OR "head trauma*" OR "brain trauma*" OR TBI OR ABI OR stroke*) AND (Peer* OR "Support Group*" OR "Group Support" OR "Social Group.")

Study selection
The author conducted the search, and reviewed all articles at each stage of the process to assess for eligibility.

Article selection summary
The systematic literature review search was carried out on 28 October 2019, and found 2816 papers, which was reduced to 2651 when the limiters were applied. The titles of these papers were then screened, and any that appeared irrelevant to the research questions were removed, leaving 79 papers. Reviewing the abstracts of the 79 papers left 27 full texts to be read. Of these, 11 papers were identified as relevant and fitting for this review. The reference lists of these papers were screened, and further two articles were identified as suitable. As such, 13 papers were included in this review, and the flowchart, adapted from PRISMA guidelines (12), in Figure 1 highlights this process.

Quality of included studies
The methodological quality of each paper was checked using the Mixed Methods Appraisal Tool (MMAT) (13). The MMAT was selected as it is can be used across different methodologies, making it suitable for the current review. The MMAT consists of two screening questions, which a paper must pass in order to be considered suitable for the MMAT. Following this there are five questions, which vary across the different types of methodology, making five the highest achievable score. In order to assess the inter-rater reliability of the scoring, a selection of papers (23%, one quantitative, one qualitative, one mixed methods) were marked independently by both the reviewer and a researcher/collegue, and reviewed for consistency of scoring. The same scores were assigned in each paper that was checked for inter-rater reliability.

Data analysis
As the papers varied in methodologies the most appropriate way to analyze the results was narrative synthesis. Narrative synthesis adopts a textual approach to synthesize evidence focusing on a wide range of questions, and identify or explain patterns and findings (14). However, the process of narrative synthesis maintains a systematic approach to searching and quality appraising data, rather than simply verbally describing it.

Declaration of interest
The authors report no conflicts of interest.

Synthesis of results
In total, 13 papers were included in this review (15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27) as highlighted in Table 1. Stroke survivors were the most frequently researched sample, and were included in seven papers, followed by five papers who recruited across a range of ABI's, and one paper who used exclusively traumatic brain injury. Sample size varied between studies from 4 individuals in one study (17) to 84 in another study (20). With the exception of Slark et al. (20), and Vandiver et al. (23), 10 studies fit within a 4-26 participant range, showing small sample sizes were predominantly used, whilst one paper did not report participant numbers. In terms of research location, the majority were conducted in the USA (8), followed by the UK (3), Canada (1), and Australia (1). Further characteristics of the papers included can be found in Table 1.
All papers broadly evaluated peer support groups; however, they all focused on different aspects, which can be separated, based on their research questions.
(1) Five papers looked explicitly at the effectiveness of peer support groups using pre vs post, or post group measures. (2) ive papers examined the experience of individuals who attended a peer support group, and three papers identified helping factors for peer support groups. For the purpose of this review, the three papers which identified helping factors for peer support groups were grouped within the experience of peer support groups.
Perhaps the most notable finding from the 13 papers reviewed was the diversity of ways that peer support groups have been facilitated following ABI. There was a variety between the papers in terms of: setting (inpatient vs. outpatient vs. community), facilitator (peer-led vs. professional led), format (fixed vs. flexible).
One key relationship between these variables seemed to be between the setting of the support groups and whether they were peer-led or professionally led. The papers reviewed highlighted that the all of the community-based peer support groups identified as being peer-led, whereas all of the groups that were established in outpatient or inpatient settings were professionally led. Additionally with regards to the format of the groups, the most notable contrast was whether they followed a fixed structure, whereby the topics or sessions were preplanned by professionals or if they were flexible, whereby topics may be offered to the group as starting points but the session content, discussions, or timetable was peer-led. No indicated relationship could be established between format and setting, for example, flexible formats were found in both community and hospital-based groups.  Stroke.
• Approximately bi-weekly sessions that were 1.5 hours in length.
• Topics set by recipients at each session, and these topics were discussed in whole group/small group discussions.
• Professional led. Therapeutic Factors Inventory. Participants reported that participation in the group was useful and they found benefits such as: helpful information, advice, making new connections and increased awareness of stroke.

Quality of included studies
The MMAT quality checklist found differences in the quality of the included studies, and trends seem to be linked to the type of methodology used. Both of the quantitative studies included were of good quality, scoring a 4 and 5, whereas the three mixed methods studies all scored a 3 or less. The reasons for the mixed methods studies scoring lower on the quality checklist were due to the absence of an adequate rationale for using mixed methods, and a lack of integration of quantitative and qualitative results. Additionally, one study used mixed methods to answer two different questions within the research and so the quantitative and qualitative data were not integrated, which is the main strength of using mixed methods (14). Of the remaining eight studies which all used qualitative methods, six studies scored a five on the MMAT suggesting they were of good quality and were appropriate in their selection of approach, extraction, and presentation of data, interpretation of results, and this was consistent throughout the research. Of the two remaining studies which scored lower on quality, similar areas of weakness were noted in relation to the interpretation of results not being substantiated by data and correspondingly, a lack of coherence between data sources, collection, analysis, and interpretation. No papers were excluded based on quality due to the aim of examining participants experience of peer support groups.

Evaluations of peer support groups using pre vs post measures
Within the four papers that offer insight into the effectiveness of peer support groups, the focus of the groups and the measures used to evaluate these varied, although they could all be described under the umbrella term of psychosocial adjustment. The four studies are varied in terms of: type of ABI, peer support group setting, length of peer support group, and structure of peer support group. That being said, three of the four studies report improvements in aspects linked to psychosocial adjustment including self-efficacy, and community integration (23,24,27). The final study examined offered mixed results, with no statistically significant findings reported (25). Across all of the studies, there were aspects of psychosocial adjustment which remained unchanged, for example, activity levels and depression scores. Two papers looked at individual's perceived self-efficacy after partaking in a peer support group. Vandiver and Christofero-Snider (23) devised a twice monthly, communitybased psychosocial support group for adults with traumatic brain injury and evaluated the impact the group had on its members using the Self-Efficacy Scale (28) and an idiosyncratic quality of life (QoL) questionnaire which asked 3 questions about present QoL, future QoL, and recommendations for services. Findings showed that compared to baseline, after 6 months participants showed increased self-efficacy scores (p < .5), suggesting that individuals experienced more perceived mastery after attendance at the group. Aspects of QoL remained unchanged and related to lifestyle/personal factors beyond the focus of the groups, such as a person's financial position. Similarly, Backhaus, Ibarra, Parrott, and Malec (29) compared a peer-directed support group, for individuals with an acquired brain injury that met for 2 hours weekly over The group ran weekly for 3 months. Each session was 45 minutes long. Professional led.
• An agenda was set in each session, but participants could bring topics of interest. An idiosyncratic stroke group questionnaire comprising of 10 statements. Participants rated each statement on a 5 point likert scale where 1 was "did not cover this", and 5 was "outstanding in covering this" Participants rated the group most favorably as a means of obtaining psychosocial support.

5/5
16 weeks, to a CBT-coping skills group. They found that participants in the peer support group showed a significant improvement in perceived self-efficacy, measured using the Brain Injury Coping Skills Questionnaire (29) after the peer support group (p < .001). The study also looked at emotional functioning and neurobehavioral functioning (caregiver ratings), and found no significant change across time. Additionally, Sadler, Sarre, Tinker, Bhalla, and McKevitt (25) reported the findings of a peer support group for stroke survivors aimed primarily at increasing resilience, although it is worth noting that this paper scored lowly on the MMAT (1/ 5).The study compared scores on the Brief Resilience Scale (30) from before the group, to the end of the group 6 weeks later, and found a marginal increase in mean resilience scores (pregroup mean = 3.6; post-group mean = 3.8). Secondary measures showed no change in activity levels, depression scores, and slight increases in physical health quality of life, mental health quality of life, and anxiety. The authors offer a possible explanation for the limited change, being that the follow up time was short and there were difficulties with missing data.
Finally, Muller, Toth-Cohen, and Mulcahey (24) evaluated how participation in a fortnightly hospital-based group impacted younger individuals who had experienced a stroke (<65). The group objectives aimed to support adjustment through health-related quality of life concepts including social, emotional, and role difficulties. As such the Stroke Impact Scale (SIS) (31) and Community Integration Questionnaire (CIQ) (32) were used at baseline and upon completion of the group, alongside a post-group survey. Results found significant positive changes across the SIS Handicap domain and both the Total CIQ score and CIQ Home Integration score, although significance was not reached across SIS Self-perceived Recovery score, and the CIQ Social, and Productivity domains.

Evaluation of the experience of peer support groups
In total, nine papers looked at the experience of peer support groups and will be outlined below. Across the studies, certain themes were noted across studies and these can be grouped under the following four themes: obtaining friendship and support (15,17,19,21,26), expression of feelings (15,16,20,26), sharing of coping strategies (15,16,18), and gaining information (17,19,20). Whilst other positive outcomes were noted, the main four themes found across studies all seem to support the use of peer support groups for ABI's.
Positive findings from taking part in peer support groups were found across a number of studies. Firstly, although its low score on the MMAT (1/5) should be noted, Pierce and Salter (15) developed a support group at a rehabilitation hospital. This group aimed to provide a safe, accepting environment to express feelings and conflicts, support problem solving, and promote help-seeking behavior. Outcomes were noted as: members developing friendships and a network of support, encouragement of problem sharing and solving within the group, shared expression of feeling, and sharing of coping strategies. Similarly, Slark, Makahamadze, Catangui, Stear, and Amorim (20) developed a monthly support group (6 sessions) aimed at 'encouraging stroke survivors, their carers, and family members to confidently enjoy life after stroke through the provision of information, education, advice, and support'.
Group evaluations were completed at the end of each session and themes were identified. Participants found it beneficial to share their experiences, and get away from the ward routine. Participants praised the group sizes, as smaller groups meant they were not afraid of speaking out, the presentation of information on the slides, and receiving information from a doctor's session.
Both Cutler, Melson, Nikoloski, and Kuluski (22), and Morris and Morris (19) report positive findings from participation in a peer support group for adults recovering from brain injury. Cutler et al. (22), evaluated a peer support group that took place in an outpatient rehabilitation service. They used semi-structured interviews, conducted 1-6 months after completion of the group and identified three core themes: (1) disrupted sense of self (pre-group), (2) enhanced psychosocial adjustment through shared experience (during the group), (3) adapted sense of self (post group). Morris and Morris (19) examined how patients experienced a hospital-based biweekly group. They reported that on the whole, participation in the group was beneficial to participants, and included participants learning helpful information, making connections, and having an increased awareness of stroke. However, group processes such as upward and downward comparison were also noted, which could be unhelpful and upsetting to some individuals.
Similarly, Pasquarello (26) reported the findings of a weekly peer support group for individuals who were recovering from a stroke in hospital. After 3 months, an evaluation was completed to measure the group against its objectives to (1) provide information, (2) offer psychosocial support, (3) offer assistance information, and (4) promote lifestyle change. The evaluation asked participants to read statements such as 'learn about the causes of stroke' and rate on a 1 to 5 likert scale as to how well the group covered that area. The group was rated most favorably as a way to obtain psychosocial support, for example, by sharing feelings and meeting other stroke patients.
Further insight comes from Oehring and Oakley (18) who reported the findings from a survey of community-based stroke support groups for younger stroke survivors (<65). Participants identified a number of unique problems to having had a stroke younger (e.g. an interrupted career), and the majority stated that they preferred a discussion format where issues could be talked over, as all participants reported feeling as though they came to help each other. When asked about this futher, a number of participants reported finding it difficult to understand speakers, and stated they would like to have discussions around relationships after stroke, returning to work, and how to 'survive by yourself'.
With a slightly different research focus, three papers sought to directly identify helping or important factors. Schwartzberg (16) completed an ethnographic study of a fortnightly peer support group for individuals who had experienced a head injury, by embedding herself within the group as a participant observer. She summarized group experiences and processes into the following ten helping factors or themes: (1) telling others about one's own pain and suffering, (2) actively listening to familiar pain and suffering in others, (3) accepting that there is a problem with group recognition of the problem, (4) grieving and laughing about daily situations, (5) receiving validation from others similar experiences, (6) being accepted by others and not having to hide one's disability, (7) supporting the survivors survival, (8) giving and receiving practical suggestions, (9) receiving and giving information from personal experiences, (10) distinguishing head injury problems from normal problems. Schwartzberg (16) concluded that the theme of legitimization and acceptance is important in understanding the findings, and that the findings support Lieberman's (33) four necessary conditions for a group to be perceived as helpful: cohesiveness, saliency, cognitive restructuring, and diversity of experiences.
Schulz (17) conducted a follow up study of Schwartzberg's (16) research using semi-structured interviews to determine participant's perceptions of helping factors, and if these differed from the ones identified by Schwartzberg (16). Schulz (17) identified 11 helping factors which can be summarized as participants benefiting from connection, support, and learning as a result of interacting with others who share the same problems. Schulz's (17) findings support Lieberman's (33) core conditions, and Schwartzberg's (18) results, with the exceptions of some helping factors such as (1) socializing, (2) finding out about other perspectives, (3) learning about others limitations and strengths, and (4) hope, not being present in Schwartzberg's (16) study.
Additionally, Tregea, and Brown (21) used focused ethnography to understand and interpret helping factors in a peer support group for individuals with aphasia using observations, interviews, and focus groups. The results highlighted 5 key themes required for an established aphasia peer support group, including (1) friendships, (2) informality, (3) a supportive communication environment, (4) providing support, and (5) the right time and place.

Discussion
The aim of this systematic literature review was to investigate the evidence for the use of peer support groups after an ABI, specifically with regards to (1) the effectiveness of peer support groups, and (2) the experience of peer support groups. To the authors knowledge, this area has not been reviewed, and so the findings from this review could offer new insight.

Summary of evidence
This search found extensive heterogeneity within the research papers across approach (quantitative vs. qualitative vs. mixed methods), methodology (interviews vs. surveys vs. measures), age range of participants (adult: 18-65 years vs. older adult: 65 + vs. mixed age: 18+), type of ABI (TBI vs. mixed ABI vs. stroke), focus of peer support groups (psychosocial adjustment vs. general support and information vs. specific factors like resilience).

Evaluations of peer support groups using pre vs. post measures
On the whole, the results suggest participation in a peer support group can lead to positive changes post-ABI. Explanations for the absence of any significant results from Sadler et al. (25) could be due to (1) missing data, and (2) the group only being run for 6 weeks. The research documenting significant positive changes are all noted to involve groups which have taken place over longer periods of time (16 weekly sessions, bi-weekly for 6 months, 9 times over 18 weeks), and it could be that the length of a group intervention is influencing the effectiveness of the group.
Interestingly, Muller et al. (24), measured perceived selfefficacy as a secondary measure using the SIS and did not find positive changes in perceived efficacy score, as previously found (23,27). One explanation for this could be due to the way the groups were facilitated. Significant improvements to self-efficacy were found in groups that used less structure, and emphasized that group structure, e.g., topics of conversation were peer-led, whereas Muller et al.'s (24) peer support group was documented to be more structured and prescriptive (although this was based on a previous survey with stroke patients). The process of being encouraged to shape the group and its content may be an empowering experience for participants, which could explain the differences in perceived self-efficacy documented within this research.

Evaluation of the experience of peer support groups
Whilst there is variation in the outcomes experienced by participants in peer support groups, the experiences seem largely positive. As highlighted in the results section, the common themes across studies can be summarized under four broad themes: obtaining friendship and support, expression of feelings, sharing of coping strategies, and gaining information. Although the studies varied in their research aims, for example, uncovering helping factors compared to exploring the experience, the findings from both groups of paper were similar in the four aforementioned themes. This suggests that the positive outcomes of peer support groups and the helping factors of peer support groups are not mutually exclusive areas. It is most likely that the ingredients needed for a successful peer support group such as it being a supportive environment, in turn end up being something that participants value and report as a positive outcome, for example, 'receiving support.' Finally, one paper did note that downward and upward comparison did occur in the groups (19), which could be a negative experience for participants, although this was not noted across other studies.
These findings seem consistent and offer further evidence to previously evidenced positive outcomes of peer support groups, which include: sharing problems and gaining mutual support and help, and gaining information (7) and broadening social networks (4,8).
These factors being identified as helpful and positive is fitting with the difficulties reported by individuals post ABI. As previously mentioned, changes after an ABI can include: difficulties with relationships, changes in activity participation, and loss of role (1). Taking part in a peer support group could provide opportunities to relieve some of these difficulties, as individuals have the opportunity to connect and socialize with others, to share, contribute, and help others in the group, and by also giving individuals the opportunity to get out and partake in the group and associated activities.
Interestingly there were a number of differences between the settings and structure of the peer support groups, but at face value, this does not seem to have impacted on the positive outcomes reported by participants. It seems that simply being surrounded by those with similar difficulties in a safe, contained environment could be the catalyst to promote positive experiences for individuals after an ABI. Although the studies are limited by small sample sizes, the promising results suggest the future investigation is warranted.

Methodological limitations and future research
Perhaps the most notable limitation of this review relates to the papers used, and the variety in: the format, setting, aim, length of peer support group, and population used. Although the findings on the whole are positive in relation to the use of peer support groups following ABI, it is difficult to make comparisons across papers, to decipher what the most effective or useful peer support group format may be. At present, there are a number of unanswered questions around factors such as the length of peer support groups and how changing these factors influences participants' experiences or the effectiveness of the group. Additionally, there are no studies comparing participation in a peer support group to a control group of participants who do not access any psychological support post-ABI. This would help establish if the positive effects noted in this review are truly from partaking in a peer support group, or may have occurred over time anyway. Future research should focus on establishing a framework for peer support groups that hold the most success following an ABI. This could include using control groups to establish if there is a need for peer support groups following ABI and where this need lies, for example, during recovery in a hospital or following discharge in the community.
Additionally, there are methodical limitations to this review that should also be noted. This includes the fact that some of the papers reviewed included family members or carers in their samples. Whilst a threshold of 50% was used to ensure that the majority of data included was from individuals who had experienced an ABI, it is likely that the peer support groups had a different impact on family members or carers than those with an ABI, and this may have influenced the results. It is also worth noting that this review was not conducted in accordance with PRISMA (12) methodology, which is the preferred way to conduct systematic literature reviews. Future reviews should adhere to this protocol, in order to ensure the clarity and transparency of reporting of systematic reviews.

Social and clinical implications
In the national context of an overstretched health-care system (28), peer support groups could be a low-cost way of addressing a sought after need by ABI survivors, and help facilitate with adjustment. National guidance suggests that everyone who has experienced an ABI should be offered psychological support (29), however research has noted that there is an increased demand for services and a lack of available resources (30). This can result in inadequate or no service provision, and/ or access to services being dependent on where you live (31). That being said, it is important that ABI survivors access specialist support, that should be not be thought off as replaced by the use of peer support groups. Moreover, whilst health-care systems could offer peer support groups, these should be used to compliment other specialist provisions delivered by suitably trained individuals.
As such, peer support groups could be a way to increase access to support after an ABI or manage waiting lists, especially in places where services are not as readily available. The variety between the peer support groups reported in this paper evidence that these groups can be adapted based on the needs of specific populations and services, for example, a peer support group could be established by staff to begin with and then later progress to being a member or volunteer run group. In many of the papers cited in this review, the 'professional' running the group varied between trainee/assistant psychologists, nurses, social workers, and occupational therapists, showing that an array of professional backgrounds could be suitable in establishing groups. It is important to note that trainee psychologists or students should not be overly relied on, and a dedicated member of staff should be involved in running the group to ensure that it remains sustainable when placements end. Furthermore, the cost effectiveness of peer support groups should be considered in relation to the larger number of participants that can be reached, rather than utilizing temporary members of a service such as trainee psychologists.

Conclusion
This review is the first to examine the use of peer support groups in ABI's, focusing on both the effectiveness and the experience of such groups. On the whole, it appears that peer support groups are associated with a range of positive outcomes. The evidence for the effectiveness of peer support groups is mixed, and this could be linked to variations in the structure and length of the peer support groups, although positive effects on psychosocial adjustment have been reported. Additionally, the research into the experience of peer support groups and the important factors within peer support groups have been largely positive, with key themes being noted in relation to sharing, problem-solving, connecting, and socializing. Peer support groups could fulfil crucial needs for individuals with an ABI, such as the opportunity to work through and problem solve issues, as well as feel empowered, offer a new role, and connect and build relationships with others in a similar situation.

Declaration of interest
The authors report no conflicts of interest