Overview
What was already known:
- Peer support has been used in prisons in England and Wales as a vehicle for easing the pressure arising from the continuously growing social care needs in social care of prisoners.
- In the long term, these peer support schemes may be cost effective since prisoners who provide support operate with little to no cost.137,138
- However, the current available evidence cannot provide clear answers about the cost of peer support schemes in adult prisons.
- Data are not available for many providers and for those where it is available they are not in a consistent format.
What this chapter adds:
- This chapter attempts to contribute to the understanding of the cost components of peer support schemes for social care in prison.
- In addition to this, it presents a cost template that can be used for future cost evaluations.
- This template identifies the key cost components of peer support schemes in prisons for social care and emerged from the workshops and literature review.
Introduction
Although many studies to date have explored the delivery of peer support schemes addressing a range of issues, there is prominent evidence gap in terms of their cost evaluation. A rapid systematic scoping review that we conducted (see Chapter 3) revealed only limited research around the cost of peer-supported social care programmes. Moreover, to the authors’ best knowledge, there are no reports or routinely collected data that could be used to explore the costs of peer-supported social care programmes.
Estimating the costs of peer support schemes in prisons is important. First, this information would allow us to know how much each scheme costs to the prison and could also help to inform which peer support initiatives are most feasible, or cost-effective, for supporting social care provision in prison and which approaches work best in different contexts. Second, estimation of costs can be used as a component in the assessment and cost-effectiveness of the schemes, as discussed in previous chapters. This is crucial given limited resources and the increasing needs of social care for people in prison, since cost evaluation and cost-effectiveness research will ensure that resources are used to meet prisoners needs and will indicate whether relocation of resources is required.
As mentioned in Chapter 2, with the current study, we aimed to address the following research questions:
- What sources of data and evidence on costs of peer support programmes are available, what is the quality and completeness of these data and who can provide this information?
- What cost-related data are needed for a future economic evaluations of peer support programmes in prison, and what is the current availability of these data?
To answer these questions, we organised the research in three parts:
- Identification of the cost components of peer support programmes for social care.
- Investigation of the costs (i.e. what was the amount of money spent for each component of the programme?), their sources (i.e. how these costs were paid?) and the prisoners’ involvement in the programme (i.e. how many prisoners delivered and received the services of the peer support programmes?).
- Investigation of the current availability and quality of relevant data.
This approach allowed us to investigate the economic aspects of peer support programmes for social care in prisons and outlines a framework for future economic evaluations of such programmes in England and Wales. Also, the approach allowed for an investigation of potential variations in sources and costs between the different programmes delivered.
In the sections that follow, we present the findings of the research in three parts: identification of the cost components; investigation of the costs, their sources and the involvement of prisoners in the programme; and investigation of the current availability and quality of these data. Using the data that we collected we conducted a case study analysis, calculating the mean cost per prisoner receiving peer-supported social care.
Methods
Identification of the cost components of peer support programmes
From the systematic review and the workshops conducted, we sought to understand all the data that are linked with the peer support programme (e.g. training costs and running costs) as well as the specific characteristics of people’s involvement in peer support schemes, including the number of prisoners providing and receiving peer-supported social care, the number of staff involved and the time spent on this by all participants. Details about the methods we used to identify the cost components of peer support programmes are given in Chapter 2.
Investigation of the costs, their sources and the prisoners’ involvement in the programme
Utilising findings from the review (see Chapter 3), workshops (see Chapters 7 and 8) and qualitative interviews with national stakeholders, prison leads and staff (see Chapters 5–7), we developed a cost template that we shared with the contacts at 18 prisons. Details on the methods we used to do this are given in Chapter 2.
Investigation of costs and prisoners’ involvement: availability and quality of the data
The cost template included questions about the existence of any data relevant to the peer support programme, and whether prisons would be willing to share these data with us. Prisons were also asked to list any costs associated with the peer support scheme and who paid for them.
For prisons that consented to share their data, we asked them to identify all costs associated with the peer support scheme between April 2022 and March 2023. They were also asked to provide information on a series of questions about prisoners’ participation, the characteristics of their involvement, and the costs related to these programmes.
To calculate the cost of prison staff involvement in the programme, we assumed that the unit cost for a staff member in salary band 6 was £58.70 per hour, and in salary band 3 it was £18.40 per hour.139 Discounting was unnecessary, because we measured costs over a short period of time (1 year). The reference year of the costing calculations was 2022–3.
Findings
Identification of the cost components of peer support programmes
We identified the list of cost components shown in Table 13. There are two categories of costs: (1) labour costs (i.e. costs due to paid working hours) and (2) non-labour costs used for the peer support programme. The former category included the cost of prison staff to establish and run the service, as well as the cost of payments or available incentives given to prisoners providing peer support. The latter included costs for training booklets, uniforms, handouts, badges, polo shirts and sweatshirts, and compacts (prison contracts) that were used during the provision of peer support.
| Cost components linked with the peer support | |
|---|---|
| Labour costs | Non-labour costs |
| Cost of prison staffa and/or staff from other organisationsb to establish the service | Costs for training booklets, uniforms, handouts, badges, polo shirts and sweatshirts, compacts (prison contracts) or any other material used in the establishment of the programme or during its provision |
| Cost of prison staffa and/or staff from other organisationsb to run the service | Costs for training prisoners providing support |
| Cost of incentives given to prisoners providing peer support (typically money or vouchers) | |
We identified three groups of people involved in peer support schemes for social care: prisoners, prison staff and external organisations. Their roles are presented in Table 14.
| Group | |||
|---|---|---|---|
| Prisoners | Prison staff | External organisations | |
| Role | Provision of peer support | Liaising with local authorities | Supporting peers |
| Receipt of peer support | Supporting peers | Provision of training | |
Investigation of the costs, their sources and the prisoners’ involvement in the programme
The cost template we developed is summarised in Box 7; the complete form is in Report Supplementary Material 9.
- The number of prisoners who provided peer support for social care from April 2022 to March 2023.
- The number of prisoners who received peer support within the same period.
- The average time that prisoners who had provided peer support had spent on the programme from April 2022 to March 2023.
- Whether any prison staff had been involved in managing, training, supporting or delivering the peer support for social care service, and if so, then what was their role, their salary band, their working hours and their involvement.
- Whether any other organisations had been involved in supporting the training or the management of the peer support for social care scheme, and if so, then what was their role, their working hours since April 2022 and their involvement.
- What other non-labour costs were incurred, such us uniforms, handouts, training packs, and if so, then what was the cost per unit and who paid for these.
- Whether there were any incentives for prisoners providing peer support.
- What was the estimated cost of the programme.
Investigation of costs and prisoners’ involvement: availability and quality of the data
In total, 10 prisons out of the 18 returned the cost template (response rate 55.6%) and most of them (60%) reported that they started to run peer support schemes between 2013 and 2016. Five (50%) reported that they collected data regarding their peer support programme. Of the others, 1 (10%) did not reply whether they collected relevant data or not, and the other 4 (40%) replied that they did not collect relevant data on costs, though some data were provided about the relevant cost components.
Prisons that collected relevant data were not all using the same data collection system, meaning that different information was collected and often in different units (e.g. numbers referring at different time periods).
The five prisons that collected data showed great variation in terms of the information gathered. This included written records (i.e. minutes) from monthly meetings with peers that can be used as data source, data about the responsibilities for recruiting peers and the employment activity, data about the training that was provided to peers prior to the peer support programme, data about the number of prisoners providing and receiving support, consent forms and data about the kind of support needed, as well as data about the frequency with which support was providing. Table 15 gives details on what was collected by each prison.
| Prison ID | Data collected |
|---|---|
| A | Responsibilities for recruiting and implementing the buddies on the wingAny training given on the roleApplications and compacts (prison contracts) for the role |
| C | Peer support scheme monitoringEmployment activity data on records of the operational databaseProvided training |
| H | Tasks requiring support with (this is in an overview sheet format for each prisoner who is going to be receiving such peer support)Contract (i.e. a compact signed by each prisoner agreeing to the support)A daily occurrence sheet evidencing the support completed (at present these are paper copies) |
| M | Number of prisoners providing peer supportNumber of prisoners receiving social supportJob descriptions |
| R | Numbers of prisoners providing peer supportWho the peers care forHow often support is provided for each client |
Seven prisons (70%) listed costs associated with payments to peers providing support between April 2022 and March 2023. All of these costs were paid by the prison. The range of these payments was between £0.88 and £2.40 for each peer support session between prisoners and each session lasted for a morning or an afternoon (half of the working day). These findings were in line with the findings from the qualitative interviews. The number of sessions per week ranged from 8 to 14 sessions (available data only for 4 out of the 10 prisons). Therefore, the minimum cost per week is £8.80, while the maximum is £33.60.
Two prisons (20%) reported additional costs that included the cost of the prison officer to train prisoners to provide support, to monitor and support the services (mean £29 per week), as well as the cost of training resources for peers providing support (mean £24 per week). The prisons reported that, along with the local authority, they were responsible for these costs.
As for prisoners providing peer support since April 2022, 2 prisons (20%) reported that 18 prisoners had been recruited to provide support. Two other prisons (20%) reported that the number of prisoners providing support ranged from 20 to 30, depending on how many prisoners were requiring support, how many were transferred in and out of the prison, or being released. Three more prisons (30%) reported lower numbers of prisoners providing support, ranging between 2 and 8. One prison reported that it had a pool of trained and available peers which was 64% greater than their actual recruitment – 18 peers in the pool, with a maximum of 11 delivering support (Table 16).
| Prison ID | Number of prisoners providing peer support from April 2022 until March 2023 | Number of prisoners receiving peer support from April 2022 until March 2023 |
|---|---|---|
| A | 18 | Unknowna,b |
| B | 8 | 10 |
| C | 25 (distributed in two sites as 10 and 15) | Unknowna,c |
| H | 4 | 10 |
| I | 2 | Unknowna |
| K | Unknown | Unknowna |
| M | 20–30d | 35–45d |
| N | Unknown | Unknowna |
| O | Unknown | Unknowna |
| R | 18e | 52d |
The number of prisoners receiving peer support fluctuated from 10 [for 2 prisons (20%)] to 52 [1 prison (10%)]. The rest of the prisons (50%) reported that the number of prisoners was either unknown or varying depending on the number of prisoners transferred in and out, released, and seeking peer support. Details about each prison are given in Table 16.
These findings are in line with those reported in the interviews, reported in Chapter 5. Small fluctuations in the estimation of these numbers can be attributed to the fact that different respondents were used for these two types of data collection as well as to the different time points when this information was collected.
Table 17 presents the variation in the average time spent by prisoners providing support for social care. The estimations reflect the workload of a peers providing support in each prison among peer support schemes running from April 2022 until March 2023. The heterogeneity in these figures may reflect the extent of support provided, for example the number of prisoners receiving peer support and whether support was provided on an individual basis or in groups.
| Prison ID | Average time spent by prisoners providing support |
|---|---|
| B | 1 hour per daya |
| C | 27.5 hours per week |
| H | Approximately 16 hours per week |
| I | 22.5 hours per week |
| K | Approximately 20 hours per week |
| R | 16 hours a week |
Nine out of 10 prisons (90%) reported that prison staff were involved in the programme, with their role being focused mostly on the management of equalities. Other roles or tasks reported included safer custody, listener support, safety manager, delivering dementia awareness, carers management, facilitating meetings, liaising with local authority and administration support.
In terms of the cost of their involvement, there was variation in their salary band, ranging from 3 to 7. Significant variation was also found in the working hours of prison staff spent on peer support, up to a maximum of 16 hours per week.
Most prisons (60%, n = 6) reported that external organisations, such as charities, were involved in peer support programmes. Their role included team management, social work, provision of training, delivering courses, and provision of kinetic support, but further information about the costs of these activities was not available.
Apart from the staff costs and those related to peers providing support, as described above, prisons reported additional operating costs for the peer support programme. Those include the costs of uniforms, buddy polo and sweatshirts, costs for sterile gloves for carers when cleaning the cells or collecting meals, the costs of badges and compacts, as well as the costs of booklets, inductions, and lanyards. Using data from two prisons, these operating costs ranged from £600 to £1100 for the purchase of uniforms used in the year of interest. Four prisons, out of the 10 that returned the cost template, reported lower operating costs, between £2 and £6 per unit, where a unit might be a T-shirt, a day providing paperwork or for one induction process. However, further information that would allow the estimation of the total operating cost was not available.
Only one prison provided an estimation of the total cost of the peer support programme, reporting that the total cost of all peers’ provision of support was £833 per year. That prison reported that peers were each spending approximately 20 hours per week providing support, but the total number of peers and recipients was unknown, and further calculations about the costs were not possible.
A case study
The availability and the quality of the collected data did not allow detailed calculations across all prisons. However, in this section, we present the case of a prison that reported adequate amount of data and could be used to calculate the mean cost per prisoner receiving peer support.
Prison C reported that they were using programmes of peer-supported social care over 10 years. Their prison had two sites contributing to the programme, and two rates of pay for prisoners: a standard scheme (standard rate of pay/incentives for prisoners) and an enhanced scheme (enhanced rate of pay/incentives for prisoners). A total of 23 prisoners from both sites were involved in the provision of support of social care (15 prisoners from site 1 and 8 prisoners from site 2) at the data collection period. All prisoners providing support did so via the enhanced scheme, and were paid £1.20 per session for delivering 10 sessions every week. For the standard scheme, when available, the corresponding cost per session was £0.88). A total of 28 prisoners across both sites received peer support via the enhanced scheme.
Given that the scheme was running for 52 weeks, and assuming that all peer supporters contributed 10 sessions per week for the full 52 weeks and all those receiving peer support did so for 52 weeks, the total cost for the prisoners who provided support was £14,352 (= £1.20 × 10 sessions per week × 52 weeks of a year × 23 peers providing support). Assuming that peer support was provided on an individual basis, the cost per prisoner receiving peer support from a buddy that received the enhanced rate of pay was £513 (= £14,352 ÷ 28) (Table 18).
| Training costs | Cost of peer support per session (£) | Number of sessions per week | Number of weeks running the programme | Annual cost per prisoner providing support (£) | Annual cost per prisoner receiving peer support (£) | Scheme of requiring support | Number of prisoners requiring support in each scheme | Total cost of scheme for meeting prisoners requirements in a year (£) |
|---|---|---|---|---|---|---|---|---|
| Excluded | 1.2 | 10 | 52 | 624 | 399 | Standard | 8 | 3192 |
| Enhanced | 28 | 11,172 | ||||||
| Total | 36 | 14,352 | ||||||
| Included | 1.2 | 10 | 52 | 711 | 454 | Standard | 8 | 3632 |
| Enhanced | 28 | 12,712 | ||||||
| Total | 36 | 16,344 |
As reported by the prison, the total training cost of the 23 peer supporters was £2000, that is £87 per peer. Adding this to the cost of payments to peer supporters gives a total cost of £16,353, which increases the cost per peer-supported prisoner to £584 (= £16,353 ÷ 28).
That prison reported also 909.5 working hours for their staff, involving six staff members. The salary band was not known for all of them, not allowing further calculations about this cost component. Regarding the costs of other resources, that prison did not report neither the cost per unit nor the number of units used, even though it reported the use of booklets and sweatshirts and polos. Similarly with the costs for the members of staff, further calculations including the cost of other resources used for the scheme, were not able to be conducted.
Discussion
Key findings
We identified key cost-components, and we assessed the completeness and the quality of the available data piloting a data collection form. Only 55.6% of the prisons (10 prisons out of the 18 which received the cost template) returned the cost template and reported some data. The rest of the prisons did not reply or did not have data to share, indicating that little or no data are collected about the costs of peer-supported social care in prisons. With regard to the quality and completeness of these data, findings indicate that prisons infrequently collect data regarding when peer support services for social care launched. This might be explained by many peer support programmes started as informal support or piloting, before they become formal schemes. There were also cases where prisons did not report information about the role of the prison staff in the programmes, or any other sources that they used for the programme. It is unclear whether or not this means prison staff were included in the provision of these programmes.
To further assess the quality of the available data, comparisons between the data collected using the cost template and the interviews took place, showing also that there were inconsistencies between the information shared by prisons. This could be explained by the non-systematic way of collecting these data and the fact that each prison could collect its own data. Another possible explanation has to do with the contact that responded to the interview or the call for the cost template. We did not contact the same person and this means that maybe the knowledge about the details about peer support programmes may not have reached everyone in the prison. Additionally, we should not ignore the fact that the time period for which we asked these questions in the two sources (i.e. interviews and cost template) was not the same. For the cost template, the questions refer to a time period of a year (April 2022–March 2023), whereas for the interviews the corresponding questions refer to the current period. Finally, we should consider also the case where the prison reported that at the time of interview it had not fully developed its peer support schemes.
In terms of the research question regarding the feasibility of cost analysis and comparisons among the prisons, we showed that the calculations for the average time spent by prisoners providing peer support was challenging. This is because we did not have further information about the frequency of the peer support sessions, the number of the contacts, and whether or not each session was attended by a group of prisoners seeking support or by a single prisoner each time. The same lack of information could explain the unfeasible comparison of the average working time and the average cost for the involvement of the prison staff.
How findings relate to previous research
The rapid systematic scoping review reported in Chapter 3 found that no studies looked at the costs of peer support schemes in prisons. Therefore, this study extends previous knowledge by exploring the costs of peer-supported social care and what cost data might need to be collected in the future. The study revealed gaps in monitoring peer support schemes, in terms of the people who are involved and the cost of these schemes. These gaps lie in the identification of this key information (i.e. numbers of people and costs), and their infrequent and not systematic collection that can be used for research purposes. This fits with the review’s finding about the existence of a unique study that partially reports cost-related findings of a peer support scheme and the non-existence of studies looking at cost-effectiveness of peer support schemes. Together, these findings suggest that research about the cost of peer support schemes for social care, and peer support schemes more generally is missing.
Strengths and limitations
The study contributes to the limited research about the cost-effectiveness of the peer support programmes for social care in prison, primarily by identifying the cost components of these services.
The main limitations were the lack of data and the low response rate. In addition, we used a purposively selected sample of prisons, while the accuracy of the data we received could not be verified. Other limitations include the non-inclusion of escalation costs – that is costs incurred when health deteriorates, requiring remedial action – and the non-inclusion of details about costs that were not paid exclusively by prisons. For example, it was beyond the scope of this study to identify wider social care costs. The current version of the prison costs form was a first attempt to explore the available data and piloted the feasibility of a consistent data collection process. As such, we did not include cost components going beyond prisons’ responsibilities. We should also acknowledge that for some types of costs, for example costs related to peers’ task of meal collection, it was not possible to distinguish between the cost attributed exclusively to the peer support programme and the cost that occurs regardless of the existence of the programme. In those cases, we considered that those costs were exclusively related to the peer support programme since they were costs that would burden the prisons even in the absence of peer support programme, but they were reported as costs related to the programme by the prisons. Finally, one of the completed forms that we received from the prisons was excluded from the analysis since the prison provided information about a different peer support programme not focused on social care. This case, also highlighted in Chapter 10, indicates that there might be poor understanding of the concept of peer support programmes for social care in prisons for members of prison’s staff.
Implications
To meet the continuously growing needs in social care for prisoners, it is important to know which interventions are cost-effective. The peer support schemes may be a cost-effective option, but the current availability of data does not allow consistent cost analyses. This study bridges this gap by developing and piloting a cost form that can be used for future cost analyses. This is of particular importance for researchers and stakeholders in prisons and social care. Findings depict the current availability of data and identify gaps and challenges that need to be addressed for the development of a database that will be consistent across different prisons. The existence of this database is necessary for future cost analyses and cost-effectiveness analyses, as well as for planning future peer support programmes.
Future research
Future research should focus on two objectives: the further development of the prison cost form and the data collection process. Further development of the prison cost form should also include the costs of treating escalations in care when health worsens as well as details about costs that are paid by the prisons and external organisations. These data will allow more accurate calculations of the costs, both for prison and wider social care services, while the availability of data about escalation costs will allow comparisons between peer support and other types of support that might be available.
As for the data collection process, future research should investigate what is the most cost-effective way of collecting data. This might include comparisons of different types of data collection processes, for example routinely and survey-based collected data, or electronic and analogue records.
Conclusions
Overall, making calculations about the cost per prisoner receiving peer support was challenging due to the availability and quality of the data. The data collection (which involved piloting questions needed to ascertain key information for economic evaluation) revealed the current lack of information regarding the details of peer-supported social care in prisons and highlighted the importance of their collection for future analyses.