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

This study is a rapid review of the effectiveness of health and wellbeing interventions in prisons, and presents findings from a survey of Scotland's prisons on the extent to which these interventions are active.

6. Conclusions and gaps in the literature

The prison-based health and wellbeing interventions included in this review showed encouraging results. Evidence from the UK and international studies suggests that sport-based; horticultural; yoga, meditation, and mindfulness; art and creative; animal-based and peer-support interventions can improve physical health and wellbeing (i.e., emotional, physical and social) outcomes for people who live in prison to varying degrees. Based on the evidence review and the results from the survey presented within this report the main conclusions and key gaps in the evidence are listed below under two headings.

6.1 Further evaluations of prison-based physical health and wellbeing interventions


The evidence review showed encouraging results for each of the physical health and wellbeing intervention categories included in this report. It is promising that Scotland's prisons deliver these interventions, albeit some are more commonly implemented than others. The intervention effectiveness classification showed yoga, meditation, and mindfulness interventions as an effective health and wellbeing intervention. However, these are not being delivered across all of Scotland's prisons. Rather, sports-based and peer-support interventions were the most commonly reported across Scotland's prisons.

The quality of the evidence identified in the evidence review was mixed. When assessing intervention effectiveness, the use of a randomised controlled design which is one of the more robust forms of evaluating interventions was limited. Some studies which utilised a randomised controlled design were identified in the evidence review, particularly for yoga, meditation, and mindfulness interventions, but overall they were limited in number. However, it has been acknowledged that conducting research in a prison setting does not always facilitate the requirements to use a randomised controlled design methodology (e.g., prisoners are transferred/released resulting in sample attrition; Meek, 2012) and there are potential ethical issues regarding the control group not having access to an intervention which would potentially benefit them. When this is the case and given the complex nature of conducting research in prison, an alternative approach researchers could take to evaluate intervention effectiveness in the prison setting is contribution analysis (Mayne, 2011). This approach has been identified as particularly useful when experimental designs are not appropriate and can determine if an intervention contributed to an outcome and in what way, based on verifying solid theories of change (i.e., identifying long-term goals and working backwards from this to identify all of the conditions that must be in place and how these relate to each other in order for the goal to be achieved). The primary value of using contribution analysis in the prison setting is that it offers an approach to reduce uncertainty about whether or not changes in outcomes are a result of the intervention by increasing the understanding of why the results occurred and the roles other factors in the prison had alongside the intervention. This then allows researchers to come to a robust conclusion about the contribution the intervention made to observed results.

Gaps in the evidence:

Further evaluations of prison-based health and wellbeing interventions in Scotland would improve understanding about which are most beneficial to those who live in Scotland's prisons. For example, while peer-support interventions were delivered in a large number of Scotland's prisons, their effectiveness classification was inconclusive (i.e., there is currently insufficient evidence to make a judgement on impact). Additional evidence would be beneficial to understand the impact these interventions may have on the physical health and wellbeing of people who live in Scotland's prisons. As well as contribution analysis being a possible step forward the following is also recommended within each intervention category when it would be possible to do so:

  • Evidence concerning the art and creative and horticultural interventions, which were classified as promising, were largely qualitative. Further research which uses a quantitative methodology with a comparison group could allow for more confident conclusions as to whether positive changes in physical health and wellbeing are a result of the intervention or not. This future research is likely to have an important impact on the classification of intervention effectiveness for arts/creative and horticultural interventions.
  • Evidence for the animal-based and sports-based health and wellbeing interventions were classified as mixed. While the qualitative evidence for these interventions was promising, the quantitative evidence varied across studies. It was also identified that a number of different interventions and assessment tools were used. While having a number of different interventions provides many opportunities for intervention modification, it is difficult to compare across them and identify which components are successfully contributing to improvements in physical health and wellbeing. Using a variety of different assessment tools to assess changes in health and wellbeing also possibly explains why mixed results were observed for certain outcomes (e.g., self-esteem). Replication studies could bring together the current evidence for animal-and sports-based interventions and may positively impact on the classification provided.

Assessments of cost and cost-effectiveness of interventions were not included in the report. This has been identified as a gap in the evidence to be addressed. When further evaluations are undertaken, information about the costs and resources required to deliver prison-based health and wellbeing interventions would be particularly useful. This could provide prison and health service staff with important information about the cost-effectiveness of particular interventions (i.e., which interventions are potentially more cost-effective and meet the needs of the prison population) to support decision making. For example, although equine therapy was named as one of the interventions respondents would like to see introduced in their establishment, it is expected that this will be more costly when compared to other interventions - such as yoga, meditation and mindfulness - and which likely require less resources to deliver.

When considering which kinds of interventions might be appropriate for their prison population, prison and healthcare staff would also likely benefit from having information about differences between the outcomes achieved by intervention categories. For example, some of the interventions teach people a skill which they can practice on their own after the intervention has finished (e.g., yoga) whereas other interventions are centred around group activities (e.g., competitive sport) which an individual cannot practice on their own. Future research considering these differences may determine if benefits observed as a result of an intervention are maintained in the longer-term in prison.

On a related point, establishing whether an intervention has an impact on people's physical health and wellbeing that is sustained after they leave prison is something future evaluations could consider. While prison is often a time when people can improve their health and wellbeing, such as through more consistent access to medical care or entering treatment programmes (e.g. opiate substitution therapy), the upheaval prison leavers often experience during resettlement can often destabilise progress they have made in prison (Jones & Maynard 2013). Accordingly,

it might be more appropriate to deliver interventions which teach an individual skills they can continue to use on their own (e.g., yoga, meditation and mindfulness) closer to their release and other interventions which require group work (e.g., sport-based, animal-based or horticultural) during the main body of their sentence.

6.2 Modifying interventions to suit the needs of sub-populations


A number of studies included in the evidence review evaluated a physical health and wellbeing intervention that was delivered to young or adult men in prison. This was particularly evident for the sports-based interventions. Evidence for the possible effectiveness of sports-based interventions for women and older adults was limited. The demographics of Scotland's prison population has changed in recent years, with the number of older adults increasing (Scottish Government, 2020). In addition, the results of the survey conducted for this report highlighted that an emerging health need in Scotland's prisons is managing the health needs of the older adult population. While challenging, making modifications to sports-based interventions, which improve the health and wellbeing of women and older adults, could prove a useful addition to services to meet the specific needs of these populations. In addition, given that yoga is potentially less physically demanding and has been identified as an effective intervention, it could be beneficial for older adults in Scotland's prisons.

Gaps in the evidence:

Further research exploring how existing sports-based interventions, such as those included in this review, could be modified to cater for the needs of women and older adults, would be beneficial. One way this could be achieved would be to include these sub-populations (i.e., women and older adults) at the design stage of the intervention to identify what their particular needs are. This offers an opportunity for researchers to identify and address the most crucial modifications to pre-existing interventions and/or developing an intervention to suit a particular health and wellbeing need. Gender-responsive interventions have been developed for women who live in prison and have received positive responses (e.g., trauma-informed interventions (Petrillo, 2021)).



Back to top