Prison-based health and wellbeing interventions: evidence review and survey of provision

2. Literature on prison-based physical health and wellbeing interventions in prison

2.1 Overview of the evidence base

Evidence for the effectiveness of physical health and wellbeing interventions for people who live in prison was encouraging, with positive effects found across a range of interventions. The evidence mainly comprised of articles from the UK, USA, Canada and Australia. A significant proportion of the evaluation literature of wellbeing interventions comes from the UK and USA. Two interventions, Paws for Progress and Inspiring Change, were evaluated in Scotland.^[7]

The research evidence included a mix of qualitative, mixed methods, and quantitative studies to evaluate intervention effectiveness. While studies which use a randomised controlled design are regarded as one of the more robust forms of evaluation (Bowes et al., 2014), their application across the evidence was limited. However, using both qualitative and quantitative research provides a more rounded answer about what interventions have worked well in the prison setting. Third sector organisation evaluations and unpublished theses were also identified across various intervention categories.

The literature showed general positive effects, with some studies showing mixed outcomes. The evidence supported the effectiveness of wellbeing interventions in improving emotional outcomes (e.g., reductions stress and anxiety). Many studies which tested interventions incorporating physical activity also demonstrated improvements in physical and health outcomes, including improved cardiovascular fitness and diet. Social outcomes were less commonly reported but positive findings included the development of friendships and social skills. However, the particular mechanisms of interventions which determined the positive impact on health and wellbeing were either not reported or tested robustly in the majority of the literature. Consequently, whether observed improvements in physical health and wellbeing outcomes were the result of someone participating in an intervention should be interpreted with caution. A primary reason for this was that many of the quantitative studies did not include a robust comparison group. However, a reason for this may be due to the well-established complexities of conducting research in prisons, which may have implications for the approaches taken. This is expanded upon in Section 6 of this report.

The literature highlighted that studies evaluating prison-based physical health and wellbeing interventions were predominately delivered to young and adult men. The effectiveness of interventions on the health and wellbeing of women and older adults were less reported. There are specific health challenges identified within these populations, including mental ill-health potentially being more prevalent in women (Bartlett & Hollins, 2018) and poorer physical mobility in older adults (HMIPS, 2017; 2021b). As such, considerations about how to modify certain categories of physical health and wellbeing interventions (e.g., sport-based) to be more age and gender appropriate is an area for intervention development identified in this review.

Overall, the literature review suggested that more evaluation of physical health and wellbeing interventions in Scotland's prisons would be beneficial. As would the development of a more robust evidence base about what interventions might be more effective in prison settings would be beneficial.

There were six categories of prison-based physical health and wellbeing interventions identified for this review based on an initial scoping search of the prison-based literature. These were: (1) sports-based; (2) horticultural; (3) yoga, meditation, and mindfulness; (4) art and creative; (5) animal-based; and (6) peer-support. These categories are similar to those identified across England and Wales, (see Turner et al., 2021). The results of each intervention category identified in the literature search are presented below.

2.2 Sports based interventions

The use of sports-based interventions to improve the health and wellbeing of people living in prison was widely evidenced across the literature. Studies evaluating sports-based interventions ranged from peer-reviewed literature to third sector organisation evaluation reports and unpublished theses (the variety across the strength of the evidence was considered when classifying the effectiveness of the interventions overall).

A systematic review which evaluated the impact of sports-based interventions on the psychological wellbeing of people in prison (Woods et al., 2017) found promising results for sports-based interventions, which covered a range of sports/physical activity (e.g., football, rugby, softball, outdoor sports including sailing and circuit training). In this review, a mix of quantitative and qualitative evidence was evaluated. For quantitative studies, which compared an intervention group to a control group (who did not receive the intervention), positive effects on emotional outcomes included significant decreases in stress, depression and anxiety and significant increases in self-esteem and positive mood were reported. Consistent with this, qualitative results demonstrated that people who live in prison who engaged in sports-based interventions reported an increase in happiness and self-confidence, as well as positive effects on physical health and diet. For diligence, the studies included in the systematic review by Woods et al. (2017) which fulfilled the inclusion criteria for this evidence review were identified and evaluated independently.

Whilst most studies in the systematic review by Woods and colleagues (2017) demonstrated positive effects, a distinction was made between the types of interventions, dependent on the emphasis placed on sport within them (i.e. whether sport was the main activity or was an aspect of a wider programme of activities that made up the intervention). This discussion is similar to a typology developed by Coalter (2007) which distinguished interventions along similar lines. Dependent on the prominence of sport within an intervention, he characterised them as "sports-only", "sports-plus", or "plus sport". These are defined below and were considered in order to identify if a particular category of sports based interventions may have a positive effect on wellbeing

2.2.1 Sports-only interventions

In sports-only programmes the majority of the activities involve playing the chosen sport and wider personal or physical development is assumed to be an inherent quality of playing sport (Coalter, 2007). Often these types of schemes utilise sport as a hook to divert at-risk groups towards more positive activities (Hartmann, 2016).

A systematic review of randomised controlled design studies reported mixed findings for the effectiveness of exercise training (i.e., sports-only) interventions on the health of people who live in prison, predominately males (Sanchez-Lastra et al., 2019). The types of exercise training used – which included aerobic exercise interventions (e.g., running), combined exercise interventions (e.g., aerobic plus resistance training) and other interventions (e.g., mixed sports) - was found to have an influence on results. For example, significant improvements in depression and general mental health were found for participants in aerobic exercise intervention groups compared to control groups. However, these positive changes in emotional outcomes were not applicable in combined exercise interventions or interventions which delivered multiple sports (i.e., strength training, golf, Frisbee, hockey, volleyball, football and basketball). Improvements in physical outcomes (e.g., cardiovascular fitness and a lower fat percentage) were reported in aerobic and combined exercise intervention groups compared to control groups.

Rugby focussed sports-only interventions for people living in prison have received empirical attention. One mixed methods study, which was completed as part of a doctoral thesis by Woods (2018), evaluated "Everybody Active 2020", a six-week rugby coaching programme delivered to young men. Quantitative results showed no significant changes for positive psychological wellbeing in the short or long-term. However, qualitative results demonstrated an improvement in relationships and an increase in positive mood and reduced stress and anger. Another mixed methods study, which also evaluated a rugby training programme for young men, found that participants in the intervention group reported positive health behaviours and a sense of belonging in interviews (Welland et al., 2020). A report evaluating two rugby programmes, "UR Game" for young people and "Walking Rugby" for older adults (55+), also reported a number of positive outcomes, consistent with a number of studies above. These included improvements in confidence, self-esteem, physical fitness and health and developing friendships and trust in others (Ulster Rugby, 2019).

Another sports-only programme identified for inclusion in this evidence review was a mixed methods study which evaluated the use of exercise referral as an intervention for men in prison in Ireland, who experienced mental health symptoms. This intervention involved people being referred to a service (e.g., the prison gym) which formally assessed that person's needs. A tailored physical activity programme was then developed and the individual's progress monitored. This intervention used a range of resistance and cardiovascular exercises to meet the needs of participants (O'Toole et al., 2017). Quantitative results found a significant increase in self-esteem alongside a significant decrease in stress, anxiety, depression and anger. These results were mirrored in the qualitative results from participant interviews, with improvements in mood (e.g., reduced stress and anxiety), self-esteem and in physical outcomes (e.g., better sleep, more energy) reported.

2.2.2 Sports-plus interventions

In sports-plus programmes sports are either adapted or augmented with additional activities to address wider behaviour or attitudinal aspects, such as workshops on anger management, territoriality, and sectarianism (Coalter, 2007).

Sports-plus programmes included in Woods et al's systematic review (2017) included UK prison-based 'sports academies' (i.e., "2^nd Chance Sports Academies"). These were prison-based programmes where sport coaching, of football or rugby was used to engage young men in prison under the age of 25 in education and training. Alongside coaching elements, participants had the opportunity to engage in other activities including watching presentations from guest speakers, achieving accredited qualifications, mentoring exercises and goal setting (Meek, 2012). Outcomes from the academies have been reported across multiple publications, showing mixed results. A wide range of benefits were reported qualitatively including improvements in social outcomes, for example, participation in the academies led to participants making friends, which encouraged communication and bonding (Parker et al., 2014), and the development of peer support which continued outwith the academy (Meek & Lewis, 2014b). Improvements in emotional outcomes, such as confidence, self-esteem and feelings of achievement (Parker et al., 2014) and being more able to cope with feelings of anger and stress, as well as positive effects in diet and health, were also reported (Meek & Lewis, 2014b). However, quantitative results reported by Meek (2012) showed no significant improvements in self-esteem immediately after participation in the academies or at a longer follow-up.

A sports-plus rugby programme in the UK, "Get Onside", was also evaluated in young men, aged 18-21, in a single prison (Williams et al., 2015). Similar to the 2^nd Chance Academies, a range of activities (e.g., goal setting and basic literacy and numeracy) were available to participants alongside playing rugby. Participants had the opportunity to develop their coaching, teaching and officiating skills. Those in the intervention group (who completed the rugby programme) reported a significant improvement in their aggression (e.g., better at controlling their tempter), whereas the participants in the control group (who completed questionnaires but were not exposed to the intervention) reported a significant reduction for this outcome (i.e., their capacity to control their aggression got worse). It was found that both the intervention and control group had similar aggression scores at pre-intervention, however, the increase in the aggression score for the control group was not explained. Some possible explanations could include intervention selection biases (i.e., those who were selected to receive the intervention were chosen because they fulfilled specific criteria, such as expressing a desire to change whereas the control group were opportunistically sampled) and the difference between the groups for the nature of their offence (i.e., the majority of the intervention group were in prison for robbery or burglary whereas the majority of the control group were in prison for violence against a person). Positive effects on aggression for the intervention group were mirrored in the qualitative findings, as participants reported feeling calmer, which suggested an enhanced sense of wellbeing. However, no differences between the intervention and control groups were observed on a quantitative measure of self-esteem.

An exploratory case study evaluated an 8-week fitness coach led programme in a prison in the USA (Amtmann & Kukay, 2016). Health and fitness coaching was identified as a relatively new and under-researched area and it differs from personal training in that it allows the individuals to set goals, participate in motivational interviewing and provides education to provide the participant with a wealth of knowledge to improve their physical health and wellbeing (i.e., it utilised the sport-plus format). Personal training, however, focusses on helping individuals exercise only. The group exercise sessions included exercise stations (e.g., sit-ups) and games sessions (e.g., basketball) which were supplemented with a session on goal setting and motivation. Quantitative analyses showed that the participants made improvements in different areas of their physical health with one participant demonstrating improvements in many areas when compared to the other (e.g., weight and cardiovascular fitness). Qualitative evidence demonstrated that the participants were aware of the improvements in their physical appearance and experienced an enhanced sense of wellbeing with reductions in their feelings of stress and anger reported. This study added to the existing minimal information about this type of intervention and a larger study would be beneficial to identify if baseline differences in certain measures (e.g., motivation) explained the differences between participants in the quantitative results (i.e., why did one participant make more improvements compared to the other?).

A mixed methods study which evaluated a programme that delivered workshops to support an in cell workout programme in a male sample in the UK reported positive effects (Baumer, 2018). This intervention involved a mix of exercise (e.g., bodyweight training which can be performed in a cell) and education (e.g., goal setting and nutrition). Consistent with the above sports-plus interventions, improvements were found in physical outcomes (e.g., weight loss) and emotional wellbeing in the short-term. These findings were mirrored in the qualitative results, which reported an increase in mood, feeling less stressed and having more energy.

2.2.3 Plus-sports interventions

In plus-sports interventions, sport is used to attract individuals to participate in education or developmental programmes (Coalter, 2007). When reviewing his typology of sport programmes for efficacy, Coalter saw plus-sports programmes as the most promising model as they do not rely on, what he sees as, the one-dimensional notion of the 'power of sport' (Coalter, 2012, p.609) where sport is viewed as possessing inherent developmental qualities and the changes that these kinds of programmes promote more likely to result from the non-sporting components (Coalter, 2012).

A mixed methods study evaluated the effectiveness of the "State of Mind Sports Programme" in men who lived in a UK prison (Woods et al., 2020). Initially developed for community settings (e.g. sports clubs), aspects of the programme were adapted to for use in prisons. This included increasing the amount of content that focussed on suicide and self-harm prevention. State of Mind aimed to raise awareness of and promote wellbeing by providing information about the markers of stress and positive coping strategies through educational sessions. In addition, two case studies were provided by former elite rugby players who experienced poor mental health and had considered taking their own lives. The purpose of these case studies was to deliver five key messages, which included: "seek help/advice from someone you trust"; "it is a strength, not a weakness to seek help"; "respond to a mate who may be feeling down"; "setting achievable goals and celebrating upon achievement"; and "we are all part of a team". No significant differences between the intervention and control group on resilience or mental wellbeing outcomes in the quantitative results were reported. However, qualitative findings revealed participants felt a sense of hope and reported positive improvements in their mental wellbeing after engaging with the programme.

2.2.4 Sports-based interventions for women in prison

Most of the sports-based studies in this review have focussed on evaluating sports-based interventions delivered to men in prison, particularly young men. Fewer studies were found which have evaluated sports-based interventions for women in prison. This mirrors the limited attention that has been paid to how physical activity could benefit women in prison (Meek, 2018). There are, however, indications that sports-based interventions might also be beneficial for women living in prison.

An exercise and nutrition intervention delivered to women in prison in Canada reported improvements in physical and emotional outcomes (Martin et al., 2013). Participants were involved in the development of the intervention. Information was collected about their perceptions of fitness, gym equipment use, nutrition and how they understood fitness and health related to other areas (e.g., sleep and stress). This was then used to design the intervention, which offered group circuit classes or individual exercise plans.

At pre-intervention, participants reported objective body measures (e.g., weight). At post-intervention, participants completed a self-report questionnaire about the effects of participating on energy levels, sleep and stress and provided objective body measures. Participants reported improvements in their sleep, energy and stress levels. A decrease in weight and BMI was also observed. Open-ended questions included in the post-intervention also showed that participants reported improvements in their wellbeing, for example improved self-esteem and reduced stress (Martin et al., 2013).

A sports-only intervention delivered to women was identified in an Australian study which evaluated sports programmes across four different prisons (Gallant et al., 2015). The intervention for women was engagement in a bi-weekly softball programme. Improvements in emotional, physical and social outcomes were reported in interviews with participants. These included reduced feelings of stress and anxiety and more engagement in social interactions (Gallant et al., 2015).

In addition, a sports-plus intervention for women who lived in prison in the USA used a pedometer to motivate walking as a form of physical activity, alongside other activities (e.g., education and social support from weekly meetings with a nurse practitioner). An objective measure of weight and height was used to calculate BMI and a self-report questionnaire was used to measure resilience. A significant reduction in BMI and an improvement in resilience, which included perseverance, self-reliance and composure, were reported (Johnson et al., 2018).

2.3 Prison horticultural interventions

Prison horticulture themed interventions can encompass a wide range of activities that include a focus on promoting rehabilitation, such as improving physical health and wellbeing, improved teamwork or communication skills (Rutt, 2016). Activities involved in horticultural interventions include the maintenance of gardens/indoor plants/outdoor spaces and growing fruits and vegetables. Consistent with sports-based interventions, evidence ranged from peer-reviewed literature to grey literature which included reports evaluating the effectiveness of horticultural programmes and unpublished theses.

One programme which has received empirical attention in the UK, is "Greener on the Outside: For Prisons" programme (GOOP). This programme aimed to reduce inequalities and achieve sustainable improvements in health, wellbeing and learning outcomes for people living in prison and their families, with a particular focus on mental health, physical activity and healthier eating. Broad programme activities included growing fruit and vegetables, maintaining outdoor space(s) and/or training in horticulture. However, rather than the GOOP model encouraged prisons to tailor their specific approach to the population who live in the prison and the physical capacity of the prison environment. Examples of these different approaches included participants developing a quiet area filled with sensory planting and seating, using polytunnels to grow plants, flowers and vegetables and the provision of horticultural training (Farrier & Kedwards, 2015).

Three evaluations of GOOP have been conducted with positive physical, emotional and social outcomes reported for men and women. For physical health outcomes, quantitative results showed participants reported fewer barriers to engaging in other types of physical activity or that they were now engaging in new physical activity (Farrier & Kedwards, 2015). This was complemented by qualitative results from focus groups and interviews in which participants reported weight loss and feeling fitter (Baybutt et al., 2019; Farrier & Kedwards, 2015). Improvement in diet and nutrition was an additional health benefit identified as a result of engaging in GOOP. Quantitative evidence found that nearly three quarters of participants reported new skills for growing and cooking nutritional food (Farrier & Kedwards, 2015). This was complemented by qualitative evidence in which engagement with the programme encouraged healthier eating (Baybutt et al., 2019). For emotional outcomes, positive mental wellbeing was reported. This included improvements in confidence, self-esteem and anxiety (Baybutt et al., 2019; Farrier et al., 2019; Farrier & Kedwards, 2015). Similarly, improvements in social outcomes, for example better relationships with prison staff and peers, were also reported in quantitative and qualitative evaluations of the intervention (Baybutt et al., 2019; Farrier et al., 2019; Farrier & Kedwards, 2015).

A qualitative study, conducted as part of a doctoral thesis, assessed the role of GOOP in improving the mental wellbeing in men who live in prison (Seymour, 2019). The results closely resembled what was found in the above evaluations, with improvements in confidence, anxiety and stress alongside the development of friendships and improved nutrition via access to fruit and vegetables reported.

An unpublished systematic review by Jenkins (2016) included various horticultural programmes delivered in prisons in the USA. This included the "Master Gardener" programme" which provided horticultural activities and the opportunity for participants to gain qualifications. The evaluation of this intervention was focussed on women who lived in prison. Quantitative evidence showed that participants in the intervention and control group (who participated in other vocational programmes) reported an increase in self-esteem and life satisfaction at post-intervention. As the control group were engaging in alternative programmes, it was perhaps not unsurprising that no differences were observed between the two groups.

Another horticultural programme included in the unpublished systematic review by Jenkins (2016) was called "Greenhouse". This horticultural programme was a year-round programme involving growing plants for community spaces in the summer, horticultural classes and access to a carpentry shop in the winter. This intervention was delivered to both men and women who live in prison. Qualitative findings showed improvements in wellbeing, which included increases in self-efficacy and self-worth, improvements in happiness, self-esteem and feeling calm, and a reduction in symptoms of depression and anxiety.

The Master Gardener Programme, which sought to assist male substance users with their recovery in prison, was evaluated in a UK prison using a mixed methods approach (Brown et al., 2016). Qualitative evidence, drawn from focus groups and reflective diaries, demonstrated the physical, emotional and social benefits of the programme. Participants reported improvements in health and physical outcomes which included better sleep, diet and fitness. Emotional improvements included an increase in happiness and reduced stress.

A mixed methods pilot study evaluating a horticultural intervention, which invited women who live in prison to prepare plants to tend in their rooms and other small trees and plants for tending in a common area, reported emotional and social benefits (Toews et al., 2018). The women who took part in the intervention reported a positive impact on their relationships with each other and feelings of calm and happiness in focus groups and interviews. The positive emotional impact was mirrored in the quantitative evidence, which was measured using a scale developed by the authors titled the "Interaction with nature scale". It measured four different emotional states, for example sadness-happiness and anchored the scale ends with emojis. The women were asked to mark along a line on the scale to indicate how they were feeling regarding each emotional state at pre-and post-intervention. These results should be interpreted with caution, however, as the scale used was not a validated, standardised measure and therefore lacks reliability and validity for comparisons with other studies.

The positive emotional impact of a horticultural intervention (i.e., gardening) was also found in an older male sample in a qualitative study in a Canadian prison. Participants reported feelings of calm and pride alongside an improvement in self-esteem and self-worth. Consistent with the nutritional benefits found in the GOOP intervention in the UK, participants reported an improvement in their diet as they had access to the fresh food they grew (Timler et al., 2019).

2.4 Yoga, meditation, and mindfulness interventions

Yoga, meditation, and mindfulness are another category of intervention that has been used to try to improve the physical health and wellbeing of prison populations. When searching the literature it was identified that these interventions can overlap (e.g., yoga and meditation) and are sometimes collectively referred to as adjunctive therapies (Auty, Cope, & Liebling, 2017). The use of yoga, meditation, and mindfulness interventions in the prison setting is well-evidenced in the literature as shown by the systematic reviews and meta-analyses which have been conducted to date.

In their meta-analysis, Per et al. (2020) reported a significant reduction in anxiety and depression for individuals who completed a mindfulness intervention (e.g., body scans and sitting meditations) compared to a control group. Positive results were also found in systematic reviews which evaluated the effectiveness of yoga and meditation interventions, when comparing an intervention group to a control group (Auty et al., 2017; Wimberley & Xue, 2016). Wimberley and Xue (2016) reported significant reductions in stress and psychological distress, alongside a significant improvement in positive mood for individuals who completed a yoga intervention compared to those who did not. Similarly, Auty and colleagues (2017) reported that people who live in prison who completed a yoga or meditation intervention experienced a moderate increase in their wellbeing compared to a control group.

It was noted that in the meta-analysis conducted by Per et al (2020), some additional analyses were ran to identify if any particular characteristics (e.g., gender) enhanced the effectiveness of mindfulness interventions. Given the number of studies identified in the meta-analysis, the meta-regression used to identify the possible characteristics was ran on studies which used a pre-post design. It was found that being female, older and engaged in the intervention for longer had positive, albeit weak, effects. The frequency and duration of a yoga or meditation intervention was also considered as a moderator in the meta-analysis conducted by Auty and colleagues (2017). It found that yoga and meditation interventions, which were less frequent but delivered over a longer duration of time, had a larger although not significant effect on wellbeing outcomes compared to shorter more frequent programmes. The benefits of longer and less intensive yoga (with and without meditation) interventions on emotional outcomes for people who live in prison has been supported by individual studies conducted in the UK and Australia as described below.

A ten-week yoga intervention, which involved a standardised set of Hatha yoga poses and stretches, was delivered once a week for two hours to men and women (7% of the sample were women) across several UK prisons. Similar findings to the above systematic reviews were found: participants reported significant improvements in positive mood, perceived stress and psychological distress compared to the control group (Bilderbeck et al., 2013). Subsequent quantitative analyses highlighted that a greater reduction in stress for the intervention group was associated with higher yoga class attendance and engagement with self-practice five or more times a week (Bilderbeck et al., 2015). In addition, engaging in self-practice five or more times a week was significantly associated with a greater reduction in negative mood for the intervention group. It is important to highlight that the parent study of the UK yoga programme (Bilderbeck et al., 2013) did not report significant reductions in negative mood as a result of the intervention. However, the study which followed reported significant reductions in this outcome (Bilderbeck et al., 2015). A potential reason for this was the different statistical approaches used. In the parent study, the intervention group was compared to a control group, therefore differences between groups were of interest. In the follow-up study only the intervention group was included in the analyses, thus differences within the intervention group were of interest only.

A mixed methods pilot study of an eight-week yoga programme delivered to men in prison in Australia which taught stretches, balances, and meditation, reported mixed results for men who live in prison (Bartels et al., 2019). Although no significant differences were observed between the intervention and control groups, this was arguably expected given the nature of the study (i.e., a pilot study with a small sample size). The quantitative results for the intervention group alone, however, demonstrated statistically significant reductions in depression and stress at post-intervention. In addition, positive affect scores rose while negative affect and anxiety scores reduced. The qualitative analyses produced results which mirrored the quantitative findings, with participants in the intervention group reporting improvements to their physical health (e.g., flexibility, strength and sleep) and wellbeing (e.g., feeling calm and happier).

A qualitative evaluation of yoga delivered to a male sample in a UK prison reported similar findings to the pilot study conducted by Bartels et al. (2019). These included improvements in strength and quality of sleep (Karup, 2016). Additional physical improvements reported were reduced joint and muscle pain and decreases in stress-related physical symptoms, for example headaches. The participants also reported improvements in emotional (e.g., anxiety) and social (e.g., improved relationships inside and outside the prison) outcomes as a result of participating.

Improvements in physical and emotional outcomes have also been reported in an evaluation of a mindfulness programme in the USA that was delivered to men in prison convicted of domestic violence offences. This intervention consisted of sixteen, hour-long group sessions which focussed on using mindfulness meditation techniques, such as thought labelling, to teach participants to identify particular thoughts. Quantitative results reported significant improvements in physical and mental health for the intervention group who completed the mindfulness programme compared to the control group (Tollefson & Phillips, 2015).

In the studies included above, the samples were predominately male, however a small number of studies identified for this review focussed on the effectiveness of yoga, mindfulness, and meditation interventions for women living in prisons in the USA. A ten-week trauma-focussed hatha yoga intervention, which was developed to address the needs of individuals who may be triggered by certain words or commands and to create a safe environment, showed positive results (Danielly & Silverthorne, 2017). Quantitative results demonstrated that women who completed the intervention reported a significant reduction in their stress and depression scores, compared to the control group. While a statistically significant reduction was not found in anxiety scores between the intervention and control groups, a reduction in anxiety scores was observed for the intervention group, which suggests that the results were moving in the right direction but were not large enough to produce a statistically significant effect (Danielly & Silverthorne, 2017).

A mindfulness meditation intervention which involved women attending eight weekly 2-hour meditation group sessions also reported positive effects. Each session included an instructional and/or philosophical talk, silent mindfulness meditation and walking, yoga and a closing discussion (Williams-McGahee, 2015). Quantitative results showed a reduction in stress and anxiety. These conclusions should be interpreted with caution, however, as the sample size for the study was small (n = 6) and results were interpreted for each participant individually rather than for the group as a whole (Williams-McGahee, 2015). It was noted that no studies from the UK which evaluated the effects of yoga, mindfulness and meditations for women living in prison were identified for this evidence review. Given the rapid nature of this review it is acknowledged that there may be literature which was not identified.

2.5 Art and Creative interventions

The use of art and creative-based interventions is another means of improving wellbeing. These interventions can take many forms including music, visual art, writing and drama.

Music-based projects for people in prison have been found to have a positive effect on participant wellbeing. In the UK, "Good Vibrations" is a project which used Gamelan music (orchestra of percussion instruments from Indonesia) as a means of improving wellbeing. Groups of participants work together to learn and play Gamelan music to support the development of their composing and conducting skills. "Good Vibrations" has been found to have positive effects across a range of different prison populations, including women (Caulfield, 2015) and older adults (Wilkinson & Caulfield, 2017).

An evaluation report of "Good Vibrations" reported on the delivery of the programme in a closed prison in the south of England, and the PIPE unit (Psychologically Informed Planned Environment) at one prison in the North of England. PIPEs are specifically designed, contained environments where staff members have additional training to develop an increased psychological understanding of their work. This understanding enables them to create an enhanced safe and supportive environment, which can facilitate the development of those who live there (National Offender Management Service & Department of Health, 2012). Qualitative results found that participants reported an improvement in confidence and emotional benefits, which included feeling more calm and increased happiness alongside a reduction in anger and stress (Caulfield, 2015). A three-month follow-up also recorded improvements in social skills. Participants reported they had been able to make new friendships, continued to speak to people they had met on the project and were more open to talking to different types of people. In their qualitative study, about "Good Vibrations", Wilkinson and Caulfield (2017) also reported emotional and social benefits in an older male sample. This included the ability to better manage feelings of anger and being able to communicate with others more freely.

"Finding Rhythms" is a music programme which was evaluated using a mixed methods study in a UK prison setting (Kyprianides & Easterbrook, 2020). In this programme, men and women who lived in 13 prisons were invited to work with music professionals to create a music album. Qualitative data was only collected from two prisons. Significant improvements in emotional outcomes (e.g., self-esteem) and overall wellbeing were found in the quantitative results. These improvements in wellbeing were mirrored in the qualitative findings. Participants reported an increase in confidence and emotional benefits (e.g., managing feelings of anxiety and depression). Qualitative findings also showed improvements in relationships amongst participants which facilitated positive social interactions outwith the intervention.

An evaluation of three "Inspiring Change" art interventions delivered to young men in a prison in Scotland included two music-based interventions. "Music for Change" taught participants how to play and record music individually and as a group, with the added option to participate in a final performance. "VoiceMale" provided workshops for song writing, group singing and vocal training, also with a final performance. Qualitative findings showed positive improvements in wellbeing. This included an increase in confidence and self-esteem (Anderson et al., 2011). The third "Inspiring Change" intervention was an arts-therapy intervention which gave participants the opportunity to create a self-portrait based on their own lived experiences. Improvements in confidence and self-esteem were also reported for this intervention (Anderson et al., 2011).

The use of art therapy was incorporated into the "PLAN-A programme", an intervention which aimed to reduce gang-affiliated violent reoffending amongst young men in a UK prison. While art therapy was one component of this intervention, PLAN-A also included 1:1 mentoring and restorative justice procedures (mediation either directly or indirectly between the individual who lived in prison and the victim of the crime). The art therapy component of PLAN-A was found to have emotional benefits for participants, including the ability to deal with feelings of anger and feeling calmer. Art therapy was also found to benefit participants socially because it provided an opportunity for individuals to have access to positive peer support. Improvements in self-esteem, mental health and wellbeing were reported for the "PLAN-A programme" overall, and it was acknowledged that the art therapy may have contributed to these findings (Meek et al., 2015).

The use of writing as an intervention has been shown to improve wellbeing outcomes for people in prison. A study of a penfriend programme for men living in prison, which involved participants writing and receiving letters from trained volunteers in the community. Outcomes included relief from feelings of isolation, feelings of happiness and improvements in confidence (Hodgson & Horne, 2015). A small mixed-methods pilot study, which evaluated the use of a brief expressive writing intervention to reduce stress amongst women in a prison in the USA, found that writing about a topic of their choice for 20 minutes across five consecutive days was a helpful method for managing emotions with feelings of general relief and reductions in stress reported (Pankey et al., 2016). Quantitative results showed a reduction in stress scores from pre-to post-intervention and while one-month post-intervention stress scores had risen, they did not return to pre-intervention levels. A possible reason for this increase was that all of the women had been released from prison and had to re-adapt to life in the community at the time of the follow-up.

There are also indications that writing interventions are particularly beneficial for marginalised groups within prison populations. In their study of a creative writing intervention, called "Dreaming Inside", which was created for Aboriginal and Torres Strait Islander men living in prison in Australia by members of their own community, Hanley and Machetti (2020) found that participants reported emotional benefits. These included the ability to manage emotions, positive feelings of self-esteem and pride in their writing achievements.

The use of performance and drama-based interventions were found to have a positive impact on the wellbeing of women who live in prison. A mixed methods study evaluated a theatre intervention called "Scratching the Surface" which delivered applied theatre techniques, such as role play, in a UK prison. Quantitative results found significant improvements in overall wellbeing alongside a significant reduction in hopelessness. The women selected for this intervention were viewed as vulnerable and at risk of suicide and/or self-harm. As such, the reduction in hopelessness was particularly encouraging as it showed that the women who scored high for suicide risk at pre-intervention were at a reduced risk post-intervention. The qualitative findings from this study supported the improvement in wellbeing with an increase in confidence and self-esteem reported by participants (Stephenson & Watson, 2018).

As well as participating in drama-based interventions, watching drama performances has been found to have a positive effect on wellbeing. A comedy performance intervention which explored mental health issues and encouraged help-seeking behaviour in a UK prison found an improvement in participant's plans to look after their own mental health. Quantitative results found significant improvements in the ease at which participants felt about approaching various people who worked in the prison setting about their mental health. This included fellow prisoners, prison officers and healthcare staff. Participants also reported an increase in engaging with positive coping behaviours, with 39% of participants saying they planned to start using the gym (Wright et al., 2014).

2.6 Animal-based interventions

The use of animals in the prison setting has been found to improve social and wellbeing outcomes. A variety of animal-based interventions were identified and included in this review.

Dog based training programmes take various forms depending on the needs of the prison population. These types of interventions are found to benefit both animals and humans. A review of prison animal programmes, in which the majority used dogs, included qualitative data which showed positive emotional, physical and social effects for participants. These included feeling less lonely, improved social skills and weight loss (Mulcahy & Mclaughlin, 2013). Few studies in the review included quantitative data for wellbeing outcomes. Of those that did, participants in the intervention group reported a significant improvement in social skills compared to the control group (Mulcahy & Mclaughlin, 2013).

A meta-analysis which evaluated the use of a variety of dog-training programmes across male and female samples identified a small but significant effect for this intervention on emotional outcomes. This included improvements in depression, loneliness, self-efficacy and self-esteem (Cooke & Farrington, 2016). Consistent with these findings, a qualitative study which evaluated the effectiveness of a dog-training programme with a female sample in the USA reported positive improvements in emotional outcomes. In this intervention, women were responsible for training and caring for both service and shelter dogs. The women cared for the dogs either individually or in pairs and lived with the dogs in a small dorm. Improvements in anxiety, self-efficacy and stress were reported (Cooke & Farrington, 2015). Based on the qualitative responses provided by the women, the authors ranked them according to overall improvement. Although the majority of the women who reported improvements had been in prison for over a year and had engaged in the programme for over 6 months, one women who joined the programme shortly after arriving at the prison was also found to benefit (she was ranked 4^th out of the 12 participants by the authors). Dog-training programmes therefore may have the potential to help women at various stages of their sentence.

A dog-training programme in the USA, "Healing Species", provided training, socialisation and general care instructions for dogs. A qualitative study which evaluated this programme in two maximum-security male prisons reported improvements in emotional outcomes, such as feeling calmer, improved self-esteem and reduced stress (Smith, 2019).

Another dog-training programme identified was "Paws for Progress", which was delivered to young men in a Scottish prison. This programme taught participants how to train and care for dogs in preparation for them being rehomed. A quantitative evaluation of the programme reported significant improvements in interpersonal and social outcomes (e.g., making friends), for those who completed the intervention. However, no significant improvements in self-esteem were reported (Leonardi, 2016). The positive effect of the intervention on social relationships was mirrored in a qualitative evaluation where participants reported they could better relate to peers after struggling to do so before the programme. Improvements in self-efficacy, mood and managing emotions were also reported (Leonardi et al., 2017).

Animal assisted interventions (AAI) have also received empirical attention in the literature. AAI is an umbrella term which includes animal assisted therapy (AAT) and animal assisted activities (AAA) (Fine et al., 2019). AAT is a structured and individualised therapeutic intervention delivered by a health provider with an animal used as an integral part of the treatment (Villafaina-Domínguez et al., 2020). AAA is a less formal intervention which is not individualised, but aims to provide opportunities for individuals to educate and motivate themselves and enjoy recreational time with the animal (Villafaina-Domínguez et al., 2020).

A systematic review of AAT and AAA interventions reported significant improvements in anxiety and depression for participants in intervention groups compared to control groups. A group AAT, delivered to women in prison, used a combination of psycho-education and therapeutic intervention techniques. In this intervention the dog was used as a proxy to discuss topics. For example, the dogs' "boundaries" were used as a way of recognising the boundaries of other people. Quantitative results reported no significant improvements in social outcomes (e.g., loneliness or conflict with others) in the intervention group compared to a control group (Jasperson, 2013). A mixed methods study tested an alternative AAT, the "St. John Ambulance Therapy Dog program" which used dogs in individual therapeutic sessions in a Canadian psychiatric prison. The quantitative results found that across the small sample participants showed a significant improvement in their emotional states (e.g., happiness) at post-intervention. These emotional benefits were mirrored in interviews with the participants (Dell et al., 2019).

An exploratory study within a special unit of a UK prison, which provides intensive care and support, delivered an AAA intervention to three men. This intervention focussed on participants' interactions with different animals, including dogs, chickens, goats, ducks and miniature ponies which lived in a purpose built animal centre within the unit. The majority of the interactions were with the two dogs in the unit. Qualitative findings showed an improved mood (e.g., feeling happy) and confidence as a result of the animal interaction (Mercer et al., 2015).

While most studies demonstrated the positive effects of dogs in the prison setting, the use of an equine-facilitated learning intervention, which taught young males in a UK prison natural horsemanship skills by teaching a horse games (e.g., asking the horse to move its feet), was also found to improve wellbeing. Based on qualitative findings, participants reported improvements in confidence and feeling calm (Hemingway et al., 2015).

2.7 Peer-support interventions

A systematic review which synthesised evidence on peer-based health and wellbeing interventions in prison settings found positive effects on wellbeing for those who delivered and received the intervention (South et al., 2014). Similar to sports-based interventions, a typology was developed for peer-based interventions which included peer-education and peer-support classifications. Peer education interventions involved an individual living in prison undergoing training and then acting as an educator to communicate information and encourage their fellow peers to engage in healthier and less risky behaviours (South et al., 2014). A large number of these interventions in the literature refer to the prevention of blood-borne viruses and were therefore excluded from this review.

Peer support interventions can involve an individual living in prison providing practical help (e.g., fetching meals), emotional and social support and advice to their peers (South et al., 2014). Examples of peer-support interventions which have reported improvements in wellbeing include the Peer Support Team in Canada and the Listener Scheme in the UK. The Peer Support Team trains women who live in prison to provide one-to-one emotional support to their female peers when it is requested. The Listener scheme selects, trains and support "Listeners" to provide confidential emotional support to their peers who may be experiencing distress. Qualitative findings showed that being a deliverer of these peer-support interventions improved self-esteem (although this was not observed in the quantitative results), self-worth and confidence. In addition, deliverers of a peer-support intervention recognised an increase in their knowledge and associated this with improved social relationships. They also reported a positive change in their relationship with prisoner staff. However, being a deliverer of the Listener scheme was also found to cause "burnout" and viewed as an emotional burden. For those who received a peer support intervention, particularly those early on in their sentence, reductions in feelings of depression, anxiety and loneliness were reported (South et al., 2014).

A mixed methods study, evaluating the Listener scheme across four UK prisons – and which included male, female and young people living in prison - found that those who received the intervention reported feelings of hope and relief post-intervention. However, feelings of anger and anxiety were also reported (Jaffe, 2012). "Listeners" were also found to benefit from delivering the intervention with improvements in self-esteem and self-worth reported. Furthermore, they reported that engagement in the intervention developed communication skills, which in turn had a positive impact on their personal relationships with staff, friends and family outside of prison.