Overview
This chapter draws on a manuscript by Walton et al., published in Public Health (https://doi.org/10.1016/j.puhe.2024.08.002). Reproduced with permission from Walton et al.47 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) licence, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The text below includes minor additions and formatting changes to the original text.
What was already known:
- Peer support services have been used in a range of prison settings for different purposes.
- Methodological quality of peer support schemes for offender health is poor, thus limiting effectiveness and cost-effectiveness evaluations.
What this chapter adds:
- A variety of peer support services have been used in prisons internationally to support a range of health, social care and educational needs.
- Different methods used to measure effectiveness (e.g. surveys and cohort studies), implementation and experience (e.g. interviews, surveys and observation). No studies measured cost or cost-effectiveness.
- Some positive effects of these peer support schemes but limitations due to data quality.
- Many factors (individual, organisational or service) influence implementation of peer support schemes.
- Range of challenges associated with peer support identified.
Introduction
Chapter 1 highlighted that adult prisoners require both professional and peer support to address a range of health, social care and psychological support needs,2,8–11,16,38,39 but that the peer support services on offer may vary across prisons.40 To the authors’ knowledge, the effectiveness, cost-effectiveness, implementation and how different stakeholders experience peer support services in prisons are not adequately studied. Findings from a 2014 review of effectiveness and cost-effectiveness of peer support schemes to support offender health in prisons highlighted that the methodological quality of studies is poor, thus limiting effectiveness and cost-effectiveness evaluations.38,39 Therefore, we need to explore how best to evaluate such schemes (e.g. what types of data are used and what outcomes should be studied), and what is known about their effectiveness, cost-effectiveness, implementation, and how different stakeholders experience these services.
This review aimed to systematically evaluate available evidence on prison peer support services. In this review, we answered the following research questions:
- What outcomes and economic outcomes, if any, have been studied for peer support schemes in prisons, and using what data?
- What is known about the effectiveness and cost of peer support schemes in prisons?
- How have peer support schemes in prisons been implemented and experienced?
Methods
We followed a rapid systematic scoping review process to review studies that focused on peer support services for adults (see Chapter 2).
Findings
Selection of studies
From the searches, 4930 papers were identified from which 70 studies were included in the review5,40,61–128 (Figure 2).
FIGURE 2 Flow chart of included studies, based on the PRISMA 2020 flow diagram (from Page et al.129). For more information, visit www.prisma-statement.org/ a, Note 71 studies included, but 2 studies have been combined due to overlap. (Figure not reproduced here; see the original article.)
When making screening decisions, the agreement between researchers was as follows: for the titles (97.9% agreement), abstracts (93.9% agreement) and full texts (62.5% agreement).
Characteristics of studies
The included studies covered a wide range of settings, designs and evaluation focus (Table 2 and see Report Supplementary Material 4). Most studies were undertaken in the USA or UK (n = 53, 76%). Where it could be determined, just under half the studies included more than one prison (n = 30 out of 70, 43%). The type of prison varied. Included studies used different designs (mixed-methods, quantitative and qualitative), and the length of research study varied from ˂ 6 months to up to 4 years. Many studies did not specify the length of their study. In terms of evaluation focus, the most studied aspects were effectiveness and peer experience. Most studies focused on evaluating only one aspect of effectiveness, implementation and experience.
| Study characteristics | Number of studies (n = 70) | |
|---|---|---|
| Setting | ||
| Country | USA | 30 |
| UK | 23 | |
| Canada | 4 | |
| Ireland | 3 | |
| South Africa | 2 | |
| India | 1 | |
| Ethiopia | 1 | |
| Belgium | 1 | |
| Israel | 1 | |
| Zambia | 1 | |
| Russia | 1 | |
| Italy | 1 | |
| Australia | 1 | |
| Number of prisons included in study | One prison only | 36 |
| Up to 4 prisons | 16 | |
| 5–9 prisons | 9 | |
| 10–14 prisons | 0 | |
| 15–19 prisons | 1 | |
| 20 + prisons | 4 | |
| Did not specify | 4 | |
| Type of prison | Male | 32 |
| Female | 15 | |
| Male and female | 15 | |
| Did not specify | 8 | |
| Design and method | ||
| Design | Qualitative | 30 |
| Quantitative | 21 | |
| Mixed methods | 19 | |
| Length of research study | Up to 6 months | 10 |
| 6 months–1 year | 10 | |
| Up to 2 years | 8 | |
| 2–4 years | 5 | |
| Did not specify | 37 | |
| Scope of evaluation | ||
| Evaluation focusa | Effectiveness | 32 |
| Cost/cost-effectiveness | 0 | |
| Implementation | 14 | |
| Staff experience | 16 | |
| Peer experience | 37 | |
| Recipient experience | 20 | |
| Number of aspects evaluated (effectiveness, cost, implementation, staff experience, peer experience, recipient experience) | All six aspects | 0 |
| Five aspects | 1 | |
| Four aspects | 4 | |
| Three aspects | 12 | |
| Two aspects | 9 | |
| One aspect | 44 | |
No studies evaluated costs. In one paper, costs were alluded to in the Discussion section, but we excluded that because costing was not stated among the aims of the study and the authors did not describe the methods for how the costs were derived.
Critical appraisal of papers
The critical appraisal highlighted that studies included in the review were of mixed quality. For the 49 qualitative studies, most of the included studies met the necessary criteria (criteria 1.1–1.5). For the six randomised controlled trials, there were some reporting omissions, for example around blinding (criteria 2.4) and comparability of groups at baseline (criteria 2.2). However, reporting for non-randomised studies (criteria 3.1–3.5) and quantitative descriptive studies (criteria 4.1–4.5), was more variable (due to a lack of reporting or insufficient information to determine quality). While many of the mixed-methods studies justified their design and integrated data to adequately answer research questions, again there were some reporting omissions which made quality difficult to determine (see Report Supplementary Material 5).
Findings
Summary of peer support services
Most of the studies focused on evaluating just one formal peer support service (n = 43). However, some evaluated multiple formal peer support services (n = 11) or evaluated peer support as part of a wider evaluation (n = 8). Six studies did not specify. Two studies evaluated informal peer support. Evaluations of peer support services were designed to target a range of different topics, including health education and prevention (e.g. for HIV, tuberculosis, hepatitis C) (n = 20), self-harm and suicide prevention (n = 9), emotional support (n = 5), disabilities and social care (n = 4), substance abuse (n = 4), and end-of-life care (n = 4) whereas other services did not target a particular topic (n = 13). Many different peer support services/initiatives were evaluated, with the most frequent being: The listener scheme (n = 12); the AIDS, Counselling and Education (ACE)/Counselling, AIDS, Resource and Education (CARE) programmes (n = 6); insiders (n = 4); peer support team (n = 4); Shannon trust mentors (n = 3), therapeutic communities (n = 3) and buddy services (n = 3). Most programmes were delivered in a group format (n = 23), one to one (n = 19) or both (n = 2). Nine studies were not clear, and this was not applicable for one study. The type of support provided by peers varied, with the most frequent types of support being emotional support (n = 28) and peer education (n = 24) (Table 3).
| Type of peer support provided by peers | Number of studies | Number of studies quantitatively measuring effectiveness |
|---|---|---|
| Emotional support | 28 | 7 |
| Peer education | 24 | 19 |
| Social care (‘buddies’) | 8 | 0 |
| Peer mentoring | 8 | 2 |
| Discussion (through support group) | 8 | 4 |
| Logistical support (e.g. induction and housing) | 5 | 0 |
| Training others to become peers | 3 | 0 |
| Direct prisoner health care | 5 | 1 |
| Screening and referral | 7 | 6 |
| Reading and literacy | 2 | 0 |
| Health trainers | 2 | 1 |
| Peer navigators | 1 | 1 |
| Mediating | 1 | 0 |
| Advocacy | 1 | 0 |
| Not specified/not enough information | 12 | 5 |
RQ1. Measures used to evaluate peer support schemes
Effectiveness
Of the 70 papers we included, 32 presented quantitative measurements of effectiveness. Of these, 21 reported outcomes only for prisoners receiving peer support, 8 reported outcomes for prisoners providing support and a further 3 reported outcomes for both groups. Three studies had more than one paper associated with them: for two of these we counted each paper separately as they presented different findings and for the other, we combined both papers as one since they reported the same results. The range of outcomes together with the data used to measure them are presented in Table 4. Studies reported a mix of subjective and objective measures, sometimes within the same study. Subjective measures include the impact on their lifestyles and attitudes, how they engage with services and their mental health which have been obtained by surveys or questionnaires. These include a wide range of existing tools measuring different aspects of physical and mental health, behaviour and personality (see Report Supplementary Material 6). Objective measures include the impact of peer-supported screening on the detection of infectious diseases, the impact on self-harm, suicide and recidivism where data have been extracted from routine prison or clinical records. But in some studies, surveys or questionnaires were used to measure recidivism. Only one study measured attitudes and lifestyles objectively by the impact of a peer-supported HIV education programme on the future prevalence of the disease.
| Outcome area | Measured outcome | Data sources | Number of studies reporting improvement (total studies measuring outcome) |
|---|---|---|---|
| Impact on prisoners receiving support | |||
| Disease detection via case finding (e.g. tuberculosis, HIV, hepatitis, other infections) | Case detection | Routinely collected clinical data, surveys, blood sampling on a voluntary basis | 4 (4) |
| Knowledge, lifestyle and attitudes relating to health (e.g. HIV, tuberculosis, hepatitis C) | Knowledge of disease | Bespoke surveys and questionnaires | 7 (8) |
| Influencing beliefs and behaviours | Bespoke surveys and questionnaires | 5 (8) | |
| Engagement with care services after release (HIV) | Bespoke surveys and questionnaires | 1 (2) | |
| Influence of improved attitudes and lifestyle on disease prevalence | Routinely collected clinical data | 1 (1) | |
| Mental health and related behaviour | Prisoners at risk of self-harm or suicide | Prison records | 0 (2) |
| Incidence of self-harm or suicide | Prison records | 3 (3) | |
| Incidence of aggressive behaviour | Questionnaires and routine prison data | 1 (2) | |
| Mental health | Numerous assessment tools (see full list in Report Supplementary Material 6) | 5 (6) | |
| Self-esteem | Rosenberg’s self-esteem scale | 0 (3) | |
| Perceptions of prison environment and prisoner interactions | Correctional Environment Status Inventory (CESI) and sociometric tests | 2 (2) | |
| Recidivism | Routinely collected data and surveys | 0 (2) | |
| Impact on prisoners providing support | |||
| Impact on activity post- release (e.g. recidivism, employment) | Surveys | 1 (1) | |
| Mental health, quality of life, and emotional crisis intervention | Guttman self-esteem scale, sociometric tests, CESI, and structured interviews | 3 (3) | |
| Impact on activity in prison (includes disciplinary infractions) | Bespoke questionnaires and prison records | 2 (2) | |
| Transferable skills and knowledge (e.g. counselling, mental health, attitudes, behaviour) | Bespoke questionnaires and structured interviews | 4 (4) | |
| Mutual peer attachment and support | Bespoke questionnaires, structured interviews, and McGill Quality of Life-Cardiff Short Form | 4 (4) | |
Studies of peer supporters themselves focused on their mental health, the value of the knowledge skills that they have learnt, and the impact on their lives after release.
Implementation and experience
Forty-nine studies explored implementation and/or experience. Many different methods were used, of which the most frequent were interviews, surveys and observations (Table 5).
| Method used to explore implementation and experience | Total number of studies (n = 49) | Implementation | Staff experience | Peer experience | Recipient experience |
|---|---|---|---|---|---|
| Interviews | 43 | 11 | 16 | 33 | 17 |
| Surveys | 14 | 4 | 5 | 4 | 9 |
| Observations | 10 | 5 | 6 | 5 | 7 |
| Focus groups | 6 | 1 | 0 | 4 | 1 |
| Notes | 2 | 1 | 1 | 0 | 0 |
| Participant journals | 2 | 2 | 0 | 0 | 0 |
| Video/audio recordings | 1 | 1 | 0 | 0 | 0 |
| Programme data | 2 | 2 | 0 | 0 | 0 |
| Informal conversations | 1 | 0 | 1 | 0 | 0 |
RQ2. Effectiveness of peer support schemes in prisons
Quantitatively measured effectiveness
A comprehensive assessment of the effectiveness of peer support programmes is difficult because of the lack of studies and the range of interventions, measured outcomes and methodologies used. Across these studies, there is consistent evidence that peer support schemes improve the detection of infectious diseases among prison populations. However, with schemes that focus more on healthy lifestyles, findings are more mixed. Prisoners receiving peer support obtain greater knowledge about their health and risks to health (seven out of eight studies), but this does not always appear to manifest in improved lifestyles or attitudes (five out of eight studies). The one study that measured the impact of a peer-supported HIV education programme on the future prevalence of the disease showed positive findings with improved viral suppression.75 This same study reported improved engagement with mental health services and fewer visits to emergency care services up to 6 months post release, but not at 12 months.
Each of the three studies of the impact of receiving peer support on suicide and self-harm indicate that, whereas there is no reduction in numbers of prisoners at risk, there is a reduction in incidence among prisoners who are at risk.77,93,106
Evidence for improvement in the mental health and quality of life of prisoners receiving peer support is variable, with most success reported with reducing anxiety and depression, and mixed success with reducing violent behaviour and self-esteem.75,78,97,98,106,111,120,125 Both studies that measured an impact of receiving peer support on recidivism (rearrest or reincarceration) showed positive results.94,113 There was notable variation between the different methods of delivering support.94 It is, however, important to note that the studies did not report information about the type of recidivism and whether reoffending relates to the same or a different crime.
Studies of the impact on those providing the peer support show several positive findings with regard to mental health, pre- and post-release behaviour, and improved knowledge and skills. However, many of these studies were of lower quality, due mainly to small sample sizes, data collection methods and a lack of comparators.
Stakeholder perceptions of benefits and challenges of peer support schemes (including interviews and surveys)
Benefits were identified for a range of different groups including prisons, staff, peers and recipients (see Appendix 1, Table 21). For prisons, the most frequently reported benefits relating to enhancing the community, including enhanced staff and prisoner relationships and sense of community (n = 5 out of 49), and a better prison atmosphere (n = 4). For staff, the most frequently reported perceived benefit of peer support schemes was increasing capacity and reducing workload (n = 11), but benefits relating to safeguarding and benefits for staff were also identified. For peers, the most frequently reported benefits related to self-development, including personal growth (n = 22), development of skills for jobs (n = 21) and motivation (n = 13). For recipients of peer support, the most frequently reported benefits were connecting with others, such as receiving support from someone similar to them (n = 16), strengthened relationships (n = 16), emotional support (n = 9) and support from staff (n = 6).
While many benefits were identified, the review also highlighted challenges associated with peer support schemes (see Appendix 1, Table 22). The most frequently reported challenges across all groups of stakeholders (staff, peers and recipients) were burden for peers (n = 12, most frequently reported from the peer perspective), confidentiality (n = 8), prisoner safety and ensuring the right motivations (n = 7), the potential for abuse (n = 9), and queries around who should be allowed to be involved (n = 5). Additionally, studies frequently reported hostility as a challenge identified by peers only (n = 8). Staff and peers also highlighted a challenge associated with peers dealing with problems beyond their role (n = 6).
RQ3. Factors influencing implementation, delivery and receipt of peer support services in prison
We identified several influential factors that affect implementation of peer support services, including stakeholder buy-in and engagement, a shared responsibility to work together, availability of training, planning and development of the programme, and funding and resources (see Appendix 1,).
A wide range of factors which were perceived to influence the delivery and receipt of peer support were identified, including peer characteristics (n = 24), working together (n = 17), staff characteristics (n = 16), resources (n = 12) and recipient characteristics (n = 12). The most frequently reported organisational factors were rules and regulations (n = 11) and the prison estate characteristics (n = 10). The most frequently reported service level factor was education, training and support (n = 19) (see Appendix 1, Table 24).
Discussion
Key findings
There is a wide variety of peer support programmes in prisons throughout the world, ranging from disease detection through to health education and social care support. Because of this there has been a range of methods used to measure their effectiveness (e.g. bespoke surveys, standard assessment tools, cohort studies and randomised trials), implementation and experience (e.g. interviews, surveys and observation). We found some evidence of positive effects (e.g. in relation to disease detection, mental health, pre- and post-release behaviour, and improved knowledge and skills), but the sample of studies in all cases is too small and the implementation of such schemes too broad to generalise these findings. No studies measured cost or cost-effectiveness. A range of factors influenced implementation, delivery and receipt of peer support schemes including individual level factors, service level factors, and organisational factors. Examples of factors influencing implementation included stakeholder buy-in and engagement and funding. Those influencing delivery and receipt included staff resources, characteristics of prisoners and staff, teamwork, prison estate rules and regulations, and education and training. A range of challenges (e.g. burden and confidentiality) of peer support schemes were identified.
How findings relate to previous research
Peer support is used internationally to try to address public health issues and inequalities in prison. The review supports previous research which highlighted that peer support services are used frequently for offender health,38,39 but provides further detail on the wide range of services peer supporters provide in prisons internationally. Findings extend previous research by demonstrating that peer support is also frequently used to provide support in other areas, such as peer education, self-harm and suicide, and emotional support. This indicates that peer support may be used to address gaps in support, and attempt to overcome inequalities highlighted in previous research.2,8–11 However, peer-supported social care services have not been explored fully in the literature, with few studies evaluating peer-supported social care identified in the review.
Previous research explored the methods used to explore peer support in health research and highlighted that methodological quality is poor.38,39 Findings extend previous research by highlighting what methods have been used to explore effectiveness, implementation, staff, peer and recipient experience of peer support services in prisons. We have found that quantitative studies are of varying quality and are generally better where sample sizes tend to be larger, such as when evaluating disease detection programmes, and poorer when samples are relatively small numbers of individuals, such as those providing peer support. Furthermore, most studies focused on descriptive analyses and did not investigate causal or moderation relationships. The role of confounders was also not investigated adequately and there was no consideration regarding different types of crimes and offences committed. The review found that no studies have explored costs of these services. Findings indicate that qualitative and primary quantitative data collections are most frequently used but that there is a limited use of existing data sets. Findings highlighted several limitations of included studies that need to be addressed to enable robust evaluation of peer support services in prisons in future.
This review adds to previous research by systematically reviewing the different types of data that have been used to evaluate effectiveness and identifying the range of findings associated with different modes of peer support. While individual studies have explored implementation and experience, this review is seminal in synthesising findings on implementation and experience across a range of peer support services in prisons internationally. This research therefore extends prior knowledge by highlighting factors which are frequently reported to facilitate and prevent implementation, delivery and receipt of peer support services. Additionally, findings synthesise benefits across a range of peer support services for prisons, staff and prisoners. However, findings highlight some key challenges that need to be addressed to support implementation in future.
Strengths and limitations
One key strength of the review is that we have taken an inclusive approach when choosing studies (e.g. definition of peer support and outcomes). It is still possible that we may not have exhaustively captured all literature on peer support. A further strength is that the review is a mixed-methods review, integrating expertise from mixed-methods researchers to comprehensively evaluate peer support services, including effectiveness, implementation, staff experience and prisoner experience.
One challenge of this review is that some studies included in the review were part of a larger study, therefore it is possible that there may be some overlap across the included studies.
As the review was a ‘rapid systematic scoping review’, and the search was conducted up until June 2022, other relevant papers (published after June 2022) have not been included. Additionally, the rapid systematic review49 and scoping review51 methodologies underpinning this review require adaptations to a full systematic review approach (e.g. not having second reviewers for all titles and abstract decisions); therefore it is possible that some papers may not have been included due to this approach.
Implications
The findings from this review can be used to inform evidence users regarding how peer support is used in prisons, and its potential in supporting the prison service to address public health issues and reduce inequalities. Additionally, the findings from the review outline benefits and challenges of peer support schemes and factors that need to be addressed to implement and deliver peer support schemes successfully. The findings on methods used and gaps in knowledge can be used to inform future robust evaluations of peer support schemes in prisons.
Future research
Future research is needed to explore under-evaluated areas, including peer-supported social care. Findings highlight a need to robustly evaluate effectiveness and cost. While many studies focus on the views of those prisoners providing peer support, further research may also be necessary to continue to explore views of staff and recipients, and implementation.
Conclusions
Peer support services are used internationally to support attempts to address public health issues and supplement care in prisons. While many evaluation methods have been used to explore peer support schemes, robust methods (particularly for effectiveness and cost) are needed in future. Findings indicate that peer support services may have some positive impact on outcomes such as improving detection of infectious diseases and raised knowledge of health risks, as well as many benefits for prisoners, staff and prisons. However, the cost of services is unknown, and some challenges need to be overcome to support implementation.