A Pilot Programme to Facilitate the Use of Mental Health Treatment Requirements: Professional Stakeholders’ Experiences

Mental Health Treatment Requirements (MHTRs) have been available in England and Wales since 2005 but are rarely used, despite high rates of mental health problems amongst offenders. In 2018, a new protocol to facilitate the use of MHTRs was piloted in five sites in England. Aims: Understanding the experiences of professional stakeholders and identify barriers to use MHTRs. Methods: Semi-structured qualitative interviews were conducted with thirty-eight professional stakeholders and thematic analysis applied. Results: Interviewees were positive about the content and implementation of the new protocol. Interviewees described key benefits as increasing options in community sentencing, addressing a gap in service provision and facilitating offenders’ access to other services. Challenges described, included multi-agency working, sustainability of funding and the range and complexity of needs of offenders receiving MHTRs and the variation in their motivation to engage. Success factors described were having a strong steering group, staff dedicated to the project and being able to provide a broad range of support to meet offender needs. Conclusion: The MHTR pilot protocol was generally well-received and appeared to address previous barriers to the use of MHTRs. Future work is needed to evaluate the effectiveness of MHTRs and the experience of offenders who receive them. in case studies the of this pilot on the views of we recommend further research fo- cussing on service users’ and carers’ experience(s) of MHTRs through interviews and ques-tionnaires. Their perspective give additional insight into the perceived successes. Whilst discussing barriers to the use of MHTRs, the exploration of ways in which families and carers could contribute in identifying and assessing the suitability of individuals needing MHTRs is suggested. This a potential beneﬁt of their views on the pilot and could explain reasons behind the low level of take up of MHTRs and other community sentence treatment requirements (CSTRs): despite the high level of take up of CSTRs


Introduction
The prevalence of mental health problems amongst people engaged in the Criminal Justice System is high, with one in every seven prisoners worldwide suffering from major depression or psychosis and one in five reporting substance abuse (Fazel et al., 2016). Whilst it is unclear to what extent incarceration causes or exacerbates mental disorders, prison is associated with poorer outcomes including self-harm, suicide, violence and victimisation for those experiencing mental health problems (Fazel et al., 2016). Recently, there has been an international drive to reduce short-term prison sentences and increase use of community-based orders combining punishment and public protection with rehabilitation and reparation. There are comparatively few studies of the prevalence of mental health problems amongst offenders who receive community-based sentences, but probation service data from England and Wales show high levels of emotional needs, mental illness and personality disorders in this population (Fazel and Lubbe, 2005;Solomon et al., 2007;Seymour et al., 2008;Cattell et al., 2013).
Responses to the high prevalence of mental health problems amongst offenders have varied internationally, from the use of mental health courts in the USA that divert offenders from prison into monitored, intensive community-based treatment (Canada et al., 2016) to the expansion of court liaison and diversion services in other jurisdictions which assess the mental health of offenders and divert those in need into inpatient or community psychiatric care, with mixed evidence of effectiveness (Lowder et al., 2018;Albalawi et al., 2019;Canada et al., 2019).
In England and Wales, community orders or suspended sentences can be ordered as an alternative to prison time and can be augmented with one or more of twelve 'requirements', such as unpaid work or drug and alcohol treatment, to which the offender is compelled to adhere (Seymour et al., 2008). For adult offenders, a Mental Health Treatment Requirement (MHTR) can be made provided it is approved by a suitably qualified practitioner (Anon, 2012a) who confirms the presence of a mental disorder which requires and is susceptible to treatment, but not so severe as to warrant a hospital or guardianship order under the Mental Health Act 1983.
Despite high rates of mental disorder amongst offenders, the use of MHTRs in England and Wales has been consistently low since their introduction (Khanom et al., 2009;Scott and Moffatt, 2012), with <1% of community orders and suspended sentences including MHTRs in 2012 (Anon, 2012b(Anon, , 2013 and a further decline since (Anon, 2018). Barriers to their use include a lack of awareness of and limited screening for mental health problems in criminal justice settings, difficulty accessing suitable community mental health care and the need for consent (Seymour et al., 2008;Khanom et al., 2009;Mair, 2011;Mair and Mills, 2009). Evidence for effectiveness is inconsistent (Anon, 2013), yet recent mental health and offender management strategies for England and Wales have nonetheless highlighted the potential of MHTRs for addressing mental health needs within criminal justice settings (Hillier and Mews, 2018) and in 2017-2018, a new protocol to facilitate the use of MHTRs was piloted in five test bed sites in England (Long et al., 2018;. This included the provision of a new psychological intervention-based MHTR pathway for offenders with mild-to-moderate mental health needs; previously, MHTRs were generally only offered to people already in contact with or requiring secondary-care mental health interventions. In this article, we explore professional stakeholders' experiences of the MHTR pilot project in order to identify continued barriers to the use of MHTRs, and how these might be addressed.

Study design
Service evaluation employing semi-structured qualitative interviews with professional stakeholders and magistrates.
Programme: a pilot protocol to support the use of MHTRs in five test bed sites The protocol was developed in partnership between the Ministry of Justice (MOJ), Department of Health and Social Care, NHS England and Public Health England and piloted in five test bed sites: Birmingham, Milton Keynes, Northampton, Plymouth and Sefton. Detailed descriptions of the pilot protocol have been published elsewhere (Long et al., 2018;. The protocol laid out a series of aims to facilitate the provision of MHTRs (as well as Alcohol Treatment Requirements and Drug Rehabilitation Requirements, which are not addressed in this evaluation). The protocol includes improving multi-agency working and increasing availability of MHTRs, DRRs and ATRs by rapid assessment and referral, timely advice to the court, delivery of appropriate treatment and maximum waiting times (see Box 1 for full list of aims). The broad principles outlined in the protocol were operationalised and adapted at a local level through newly established steering groups in each test bed site. A programme manager working across the five test bed sites also provided detailed implementation guidance and support. All sites received a limited amount of central funding for the pilot, as well as obtaining additional local funds from a variety of agencies.
Each test bed site was required to provide a psychological intervention pathway for MHTRs, with the aim of addressing current mild-tomoderate mental health needs of service users. The content of the intervention varied between sites, but included psychoeducation, compassionfocused therapy, value-based solution focused therapy and behavioural activation. Interventions were approximately 12-week long and delivered by mental health practitioners (often trainee or assistant psychologists) under the supervision of the local clinical lead. Treatment providers included existing psychological services and third sector organisations into which dedicated MHTR teams were embedded. The mental health practitioners delivering the MHTR intervention were funded by the money that each test bed site received for the pilot.
Each test bed site was required by the protocol to identify a named clinical lead to approve MHTRs, usually a clinical psychologist. Several sites also had a dedicated assistant psychologist or similar health care professional who spent time in the courts to assist with the identification of people potentially eligible for MHTRs. This post was not mandated by the protocol but was recommended in all test bed sites by the overall programme manager. In the majority of sites, broader support was provided for those receiving MHTRs through pre-existing posts, such as link workers who supported service users with attending appointments, housing, GP registration and other needs. In Northampton, only women were eligible for the new MHTR pathway; other sites did not have eligibility criteria based on gender.

Timelines
Milton Keynes commenced their primary care level MHTR pathway in 2014, with the other four sites commencing between October 2017 and April 2018. Interviews for this qualitative evaluation were conducted between January and September 2018.

Participants
Potential interviewees were identified by the national MHTR programme manager and steering group leads from the five test bed sites. We aimed to interviewee approximately six stakeholders in each test bed site, including the steering group lead. We approached key professionals, including practitioners or managers, from mental health services, probation services, drug and alcohol services, liaison and diversion services, relevant third sector organisations, commissioning bodies and the judiciary.

Data collection
Topic guides for semi-structured interviews were tailored to each professional group and addressed professionals' roles and responsibilities related to MHTRs and their reflections on the pilot programme, its implementation and their recommendations for ongoing use of MHTRs. Interviews lasted approximately 30-60 min and were audio-recorded and transcribed verbatim.

Analysis
Transcripts were checked against audio recordings, entered into NVivo11 and analysed using thematic analysis. The analytic strategy sought to combine deductive and inductive approaches. One researcher (EM) coded the first three interview transcripts line by line and developed a draft thematic framework. The draft framework was refined following discussion with two additional researchers (N.V. and S.O.), who applied the framework to the same three interview transcripts and three additional transcripts. Subsequent interview transcripts were coded line by line by one researcher each (either E.M. or N.V.). The framework was revised following review by the NIHR Mental Health Policy Research Unit Lived Experience Working Group. Iterative revisions were made to the thematic framework as the analysis progressed. Anonymised quotes are provided to illustrate the themes generated.

Approvals
Approval for the service evaluation was obtained from the overall programme manager and steering group chairs in each test bed site. Additional approvals to interview probation staff, and members of the judiciary were obtained from the National Research Committee at Her Majesty's Prison and Probation Service (reference number 2018-036) and the Senior Presiding Judge, respectively. All interviewees provided written, informed consent prior to the interviews.

Participants
In total, we approached fifty-two potential interviewees, of whom thirtyeight took part. The most common reasons for non-participation were insufficient involvement with the pilot programme and changes to professional role. Interviews were conducted with probation services (n ¼ 9), mental health services (n ¼ 11), the judiciary (n ¼ 4), liaison and diversion services (n ¼ 4), drug and alcohol services (n ¼ 4), third sector organisations (n ¼ 2), and commissioning bodies (n ¼ 4), as well as the national MHTR programme manager and the steering group chairs in each site.

Findings
Themes were organised into three categories: benefits, challenges and success factors. The themes and main sub-themes are summarised in Table 1.

Benefits of increasing accessibility of MHTRs
Across all the professional groups, interviewees believed that there were substantial benefits associated with increasing the accessibility of MHTRs for offenders in the test bed sites. The introduction of the primary care-level MHTR pathway was seen to increase options within community sentencing and potentially reduce reliance on short-term prison sentences, which were described negatively by almost all interviewees. Several interviewees stated their belief that by providing a psychological intervention and broader support (for example with housing), the MHTR programme addresses factors which could contribute to offending behaviours and may therefore be more effective in reducing reoffending than short-term prison sentences.
They go to prison for six weeks; they've been in and out of prison for the last 10 years, is it actually going to make a change on that occasion?
The MHTR is designed to provide, as I understand it, a significant level of psychological intervention while not over medicalising the problem, while also providing the levels of social support that deal with the chaotic lifestyle issues, as well as the traumatic issues that may be causing, or part of the cause of the offending (Judiciary) The new MHTR pathway was described by interviewees as filling a gap in services. Interviewees highlighted that the mental health, substance misuse and/or social needs experienced by many in this group made it particularly difficult for them to access services in the community, but they had previously not been offered MHTRs because they did not meet criteria for secondary mental health services.
We now deal with people, not only with [people who meet the threshold for] secondary mental health [services] . . . but also primary. Which was quite a large swathe of the criminal population that we were missing out on, who obviously needed help. (Judiciary) The group that we talked about, that don't reach the threshold of secondary care mental health services, there is a commissioning gap. And if we look at the statutory services that we have, currently, within primary-care, there are very few services at the moment, that can support the needs of this, very difficult to engage, and extremely vulnerable group. (Mental Health Services) Although the 12-week psychological intervention was seen as valuable, most interviewees did not believe that it was sufficient to fully address the needs of this group, which could include, for example, trauma, adverse childhood experiences and/or sexual assault. However, they highlighted that the new MHTR pathway provided an important opportunity to increase offenders' access to ongoing mental health care or other support.
They might need a referral to a sexual assault centre or.. So, they're a hidden population that we don't really know about, so it's almost like having a golden opportunity to put people in the right place as well. (Probation Service) Finally, many interviewees described the positive feedback they had received from service users about the new MHTR pathway.
The feedback from clients is just positive, completely and utterly positive. No, it doesn't solve all of their mental health problems, but it does improve their ability to cope with life. It does improve their confidence in professionals and the mental health as a service, I guess. And also, just to give them confidence in talking about mental health, and it being a 'thing'. (Third Sector Organisation)

Challenges of the pilot
Despite interviewees' positivity about the MHTR pilot programme, several challenges were identified. Some were general challenges, such as multi-agency working and sustainable funding, whilst others were specific to the MHTR context, including concerns around service user motivation, appropriate sentencing and breach. Strategies to address many of these challenges were also described and are reported in the subsequent theme ('Success factors').
Multi-agency working was described as challenging in all sites, particularly where strong interagency relationships did not exist prior to the pilot. Interviewees highlighted the different expertise, language, frameworks, processes and goals of the different agencies, as well as their limited understanding of each other's work. There were also practical challenges, including difficulty finding suitable rooms for the treatments. For example, some mental health organisations were reluctant or refused to provide treatment spaces for people involved in the criminal justice system and in some sites, different premises had to be used depending on whether the sentence was being managed by the National Probation Service (which manages sentences for offenders with higher risk of harm) or private Community Rehabilitation Companies (which manage offenders at low or medium risk of harm). Interviewees also highlighted that there was a lack of clarity about who held overall responsibility for MHTRs.
The client or the participant, offender, defendant-everyone calls that same person a different name (Steering Group Chair) I think I've said to anyone who'd listen, it's kind of everyone's business but no-one's responsibility to make sure this exists. Is it a health job? Is it a justice job? . . .. It doesn't really matter but someone needs to be responsible for it. (Steering Group Chair) Within a multi-agency context, the pilot had to contend with competing organisational priorities. For the courts these included same day sentencing, whilst for liaison and diversion staff, these included supporting individuals with severe mental illnesses. This was particularly challenging in sites where probation and liaison and diversion services were understaffed (e.g. due to vacancies and staff absences), which was seen to have led to a reduction in the number of individuals identified for the new MHTR pathway. Interviewees reported that even when professionals were enthusiastic about the MHTR pilot, they often did not have sufficient time to devote to it.
It's about staff being pulled in many different directions. . . Just in my experience, there is never the joined-up thinking looking at the full picture. Everyone has their priorities. The pilot is one of those priorities, but then we have the other priorities to the other groups as well. (Probation Service) Relatively limited financial resources had been allocated to the pilot programmes. Some interviewees reported this meant that the pilot had to be supported through existing posts. This placed an additional burden on staff. Several interviewees also were concerned that the short-term funding for the pilot might mean that the new MHTR pathway was not sustained and its benefits would be underestimated.
When you have, maybe, 12 months' worth of money and it might take you four or five months to mobilise to get to a point where you think, "Right, okay, this is where we'd really like to be," and you've got six months left of delivery, it feels like everything, potentially, could be diluted in terms of our view of the success. So, I would really like to see, early on, some kind of commitment that helps people understand that, "Okay, we are going to go beyond a year." (Commissioner) Another issue raised by several interviewees was that the new MHTR pathway did not address the needs of those with more severe mental health problems. In most sites, the new MHTR treatments were provided by relatively inexperienced practitioners (often Assistant Psychologists), which was seen to prevent more severe or complex needs from being addressed. This was problematic as, even amongst those with primary care-level mental health needs, many service users had experienced substantial trauma or had other complexities in their presentations. Employing more experienced mental health practitioners and increasing accessibility of secondary-care MHTRs were recommended as ways to help meet these needs. Some interviewees stated that there was a need to focus more on facilitating secondary-care MHTRs in the test bed sites.
When people come in who are more complex and that have got significant trauma, that they should be seen by somebody with more experience than an assistant psychologist (Mental Health Services) I think the area that's lacking is the more serious mental health issues. We seem to deal with a lot of people who have got quite serious mental health issues, and that need isn't suitable for the psychological programme in its current form. I've got to admit that it was a source of disappointment for all practitioners, really, that it's only got limited availability. It seems to be targeted towards the lower level mental, whereas the need really lies-we think-with the more serious mental health cases. (Probation Service) However, other interviewees believed that MHTRs have limited value for people who are already in contact with mental health services.
Some interviewees viewed service users' level of motivation as a challenge to the success of the pilot. On the one hand, the compulsion associated with MHTRs was seen to help maintain engagement with services. However, there were also thought to be important negative consequences of the fact that people were required by a court order to engage as opposed to having an intrinsic motivation to do so. I think if you say to people, "Do you want a bit of therapy or do you want court?" they're going to say, "I'll have therapy," because why wouldn't you? So, I think that there's an issue around the validity of people's engagement. So how meaningful is the engagement within the therapeutic process? I think that's a challenge (Mental Health Services) Many interviewees highlighted that breach of MHTRs had been rare in the test bed sites so far but, particularly within mental health services, interviewees also expressed concerns that they or their colleagues had about enforcement and breach. They were particularly worried that individuals with mental health problems could be breached and potentially sent to prison for not attending their MHTR appointments. Some mental health professionals also had limited understanding of how breach processes worked in their area and were not able to describe the criteria for breach or its consequences. To address concerns around breach, one interviewee described how visiting another test bed site to see how breach was dealt with had been very helpful. So going and talking to the staff and understanding that, collectively, they had a really, really good threshold for breaches and you know would go the extra mile to try to contain, encourage and look at strategies that kept people connected with the programme: and that was really reassuring for me to be able to go back to colleagues and say, "Look, this is about interacting with people in the way that we normally would." (Commissioner) Finally, both perceived leniency and severity were raised as concerns about MHTR sentencing. Some interviewees working in non-judicial settings were worried that the judiciary might avoid MHTRs as an alternative to short-term custodial sentences because they did not want to be overly lenient; this was not reported by the members of the judiciary interviewed for this study. Others were concerned that MHTRs might encourage up-tariffing (i.e. disproportionately severe sentences for people would otherwise have received lesser sanctions such as fines). However, where concerns about up-tariffing were described, these were seen to have been resolved at a local level by communication and guidance about eligibility.
We just had to be really clear that although it sounds great, the risks are, the dangers are about up-tariffing sentences so don't do that because you think, "Oh a woman could really do with a psychologist's help," when all she's done has not paid a TV licence. . . (Third Sector Organisation)

Success factors
Many of the identified 'success factors' addressed the previously described challenges. First, interviewees across all sites described practical factors as having been a key to multi-agency working. These included the co-location of services, establishing information sharing protocols, and ensuring clarity about other agencies' roles and responsibilities, for example through process maps and multidisciplinary meetings. Ongoing inter-agency awareness-raising and training were also described as crucial. The establishment of regular steering group meetings in each tested site was a key, with the strength of the steering group described as vital for facilitating the implementation of the pilot. Senior and multi-agency representation was described as a key contributor to steering group strength.
If you have the right people at the table, you can influence them. Knocking on the door of the service and there's no one in a more senior position supporting that, it just makes it more difficult. (Steering group chair) Pre-existing cross-agency relationships were viewed as very important. The test bed sites were selected, in part, based on whether relevant partnerships were already in place, and some interviewees raised concerns about whether it would be possible to successfully replicate the MHTR programme in other areas if there were weaker relationships between agencies. However, many also stated that stronger inter-agency relationships had developed as a result of the pilot. Interviewees also described how multi-agency working was facilitated by the most professional stakeholders having been highly motivated to engage with the pilot. This was seen to stem from an awareness of the unmet need for access to mental health services and a good alignment between the aims of the pilot and agencies' other goals and priorities.
The grassroots pressure, combined with increasing interest by senior civil servants and ministers in sentencing policy that works rather than sentencing policy that doesn't, has led to a very fruitful coming together. . . You don't often get a situation where the academics, the deliverers and the policy makers are all saying, "We've got to do something here." (Judiciary) Several aspects of staffing were also viewed as key to the success of the pilot. Almost all sites had a dedicated member of staff (often an Assistant Psychologist) available in the court to support screening for MHTRs which could otherwise lose out to other priorities of the liaison and diversion service. However, interviewees described how support from the liaison and diversion service, as well as third sector organisations and clinical supervisors, was crucial if assistant psychologists were inexperienced in working with the criminal justice system. Many interviewees also highlighted the value of having a named clinical lead.
The crux and the key is the mental health professional [clinical lead] who gives the thumbs up to an order from court. So, it's the speed of somebody being contacted from court to say, "Will you put your name to an MHTR for this case?" And them say, "Yes," and because that's all joined up, and because there's a structure around it to support that, and I think because it's normally basically one person who knows exactly the consistency required to either agree, or not agree that. (Probation Service) Members of the steering group in each of the test bed sites had attempted to identify pre-existing statutory services that could provide primary care level MHTRs, but in all areas new treatment provision had been developed. Interviewees described this as necessary because many of the service users faced barriers in accessing and benefiting from standard statutory services, for example because they were not registered with a GP, were excluded from services due to substance use or were experiencing socio-economic disadvantages such as lack of stable housing. Shorter waiting times were described as another important feature of having specific treatment provision for MHTRs. Members of the judiciary highlighted that when they are sentencing people to MHTRs, they need to be very clear about when the treatment requirement will start, which might not be possible if accessing community services with long waiting lists. Interviewees also described the importance of providing timely support in order to rebuild trust, as it was common for those receiving MHTRs to have been let down by services multiple times in the past. Some interviewees raised the concern that MHTRs could therefore be seen to prioritise offenders for mental health care. All interviewees who mentioned this described why they believed it was justified but felt that it could be an issue for colleagues or the general public.
So I think it would be unhelpful, just, to sort of say, "Okay, they just need wider services." They need services that are much more, as you say, personalised -and that there's a greater understanding of the context of the services. Because these are often people who have. . . The context is, as much, affecting their behaviour as the complexities of their own experience. So that's another area that, perhaps, doesn't necessarily get addressed if you just go down the mainstream mental health rail. (Mental Health Services) I remember speaking to my colleague about it, and they said, "But don't you think that's a little bit" -I think they used the term unfair, that someone is offending, and gets access to that intervention before someone who isn't offending; they have to wait six months for that level of intervention. My personal view on that is, although I take on-board what they said, actually one, these people have probably had real traumatic experience, and two, in terms of cost effectiveness, the cost they're having on society in general, with their offending behaviour, and the cost of everything involved in that, is significant. (Liaison & Diversion) Finally, providing service users with appropriate support for their social needs was also described to be critical to the success of MHTRs. This included helping with housing or benefits, registering with a GP and addressing barriers that might prevent attendance at MHTR appointments. The provision of broader support was seen to enable people to benefit more fully from the psychological treatments and to help to build relationships between service users and front line professionals. This was seen to be an important aspect of rebuilding people's trust in services; as described above, professional stakeholders felt that many of those receiving MHTRs had previously been let down by services. In some sites, third sector organisations were involved in providing this broader support for the pilot programme; all interviewees who mentioned these services described them as extremely valuable.
to continue that service to provide that additional help and support, which isn't really probation, isn't really health, isn't really to do with the offending. It's more about providing practical assistance for that individual to then let them engage (Judiciary)

Discussion
Our evaluation found that, across the five test bed sites, professional stakeholders were generally positive about MHTR pilot, reporting in particular that the new MHTR pathway addressed a gap in service provision for people who needed psychological support but who did not meet thresholds for secondary mental health care. Increased accessibility of MHTRs was suggested to provide better sentencing options and improve access to mental health services. However, interviewees highlighted that there was still no well-functioning pathway for service users with more severe mental health needs.
Factors described as facilitating the use of MHTRs in the test bed sites included the establishment of engaged steering groups, the implementation of the psychological intervention MHTR pathway and the identification of a named clinical lead in each site. Clinical leads were responsible for MHTRs and supervised the health care professionals providing the psychological intervention. Interviewees highlighted challenges that can be understood within the context of the implementation of complex interventions: establishing multi-agency processes, dealing with competing priorities for staff and the sustainability of funding. Current thinking in implementation science indicates that strategies that may be most effective depend on intervention type and setting of application; there is no one criterion for introducing such interventions successfully (Powell et al., 2014(Powell et al., , 2019. Challenges specific to the MHTR context included concerns around service user needs and motivation as well as appropriate sentencing and breach. Interviewees suggested that they had been able to overcome these challenges by having dedicated staff and services for MHTRs and by providing broader packages of support for offenders. In a previous study exploring barriers to the use of MHTRs, many court professionals and magistrates reported that mental health was not the responsibility of the criminal justice system (Khanom et al., 2009). Our evaluation did not find this: the members of the judiciary and probation staff we interviewed described themselves as highly motivated to facilitate mental health support for offenders. They highlighted the high level of mental health needs amongst offenders and suggested that addressing these needs could help to reduce reoffending. Several factors may contribute to this difference in findings, including changes in attitudes towards mental illness in the decade separating the two studies (e.g. Dean and Phillips, 2015). Members of the judiciary commented in the interviews that having mental health provision embedded in the test site increased their confidence to make MHTRs as they were reassured that appropriate treatment was available. Our interviewees may also have been unrepresentative magistrates and probation staff, as many had actively chosen to be involved with the MHTR pilot. Methods of increasing or maintaining professional motivation may need to be considered if the programme is rolled out more widely.

Limitations of the evaluation
This qualitative evaluation provides a detailed analysis of professional stakeholders' experiences in the test bed sites. However, there are some important limitations. Most interviews were conducted between January and April 2018, during which time some test bed sites were only just starting to offer the new MHTR pathway. Several interviewees could therefore only suggest the impact they thought the pathway would have rather than describing its observed effect. Whilst this is a limitation, there were also benefits of interviewing people contemporaneously: challenges and success factors were described which might have been forgotten or overlooked several months later.
Whilst interviewing stakeholders involved in the pilot programme was strength for understanding the implementation of the protocol, it precluded consideration of some broader barriers to the use of MHTRs. These include, for example, potential concerns amongst health care professionals who have no experience of working with offenders. As discussed above, the enthusiasm of professionals in the test bed sites selected to be involved in the pilot might also not be as widespread in other areas, which may limit generalisability to new sites if the MHTR protocol was implemented more widely. There may be similar implications for other factors which contributed to the selection of the test bed sites, such as the strength of relevant local partnerships, particularly in Milton Keynes which had established a similar MHTR pathway prior to the national pilot. We are also not able to draw conclusions about the experiences of other relevant professional groups working in the test bed sites who were not interviewed, such as defence solicitors.
Some topics, such as the issue of consent and compulsion within MHTRs, were raised less frequently during interviews than we had anticipated but should be carefully considered within any MHTR programme. Language barriers and disabilities may also lead to challenges in accessing MHTRs but were not spoken about. These issues should not be ignored in future consideration of MHTRs. Childcare may be needed to enable service users, particularly women, who have young children to attend appointments (Grote et al., 2007). The lack of discussion of these topics may reflect the fact that we did not interview service users about their experience of MHTRs and barriers to engagement.
Interviews with service users would provide important insights to direct future developments of MHTRs but were beyond the scope of this work. Focus groups were conducted separately by Clinks (a national organisation supporting charities working in the criminal justice system) to explore attitudes to community sentence treatment requirements amongst people who have experienced the criminal justice system, although not necessarily treatment requirements specifically (Clinks, 2019). Themes from these focus groups reflected many of those identified through our interviews with professional stakeholders, including the need for personalised holistic support and the importance of relationships with professionals. Most participants with lived experience supported increasing the use of community sentence treatment requirements, as they felt these would be more beneficial than receiving a custodial sentence. Other themes from the focus groups were less central or not identified in our interviews, including the importance of individual involvement in care planning and consent to treatment, and the potential benefits of including peer support within treatment requirements.

Implications for future research
This study provides valuable insights into professional stakeholders' views and experiences of implementing the MHTR protocol in the test Having reviewed this report on MHTRs, we reflect that increasing the usage of MHTRs could be a positive move to provide better support to individuals with mental health needs at risk of reoffending as an alternative to a short prison sentence. Like many mental health services, these schemes experienced lack of resource and funding which were both identified as barriers to success. We hope that should these pilots be successful, sustainable resources would be provided to extend the schemes and support the development of a pathway for individuals who have more severe or complex mental health needs, as this is clearly lacking. In this report, service users are considered as one homogenous group, we feel the report fell short of considering how certain characteristics, such as gender, sexuality and diagnoses, may hinder or encourage commitment to the scheme. We also feel that exploring how a coordinated support network helps these groups to engage with the treatments in practice with some case studies or examples would increase the impact of this paper. As this pilot has concentrated on the views of professionals, we would recommend further research focussing on service users' and carers' experience(s) of MHTRs through interviews and questionnaires. Their perspective will give additional insight into the perceived successes. Whilst discussing barriers to the use of MHTRs, the exploration of ways in which families and carers could contribute in identifying and assessing the suitability of individuals needing MHTRs is suggested. This highlights a potential benefit of seeking their views on the pilot and could explain reasons behind the low level of take up of MHTRs and other community sentence treatment requirements (CSTRs): 'However, despite the high level of need, take up of CSTRs is very low' (Khanom et al., 2009;Scott and Moffatt, 2012).
bed sites, but further quantitative and qualitative research exploring the MHTR protocol is required. In particular, it is crucial that the effectiveness of the new primary care level MHTR pathway is evaluated to explore whether this improves mental health or offending outcomes, as this is currently unknown. Professionals interviewed for this study were clear that there has been a treatment gap for people who require support for psychological and social needs but do not meet criteria for secondary mental health care. However, further work is required to explore whether this support should be provided through MHTRs, as examined in the test bed sites, or through voluntary referrals to tailored primary care or IAPT services. Quantitative comparison studies and qualitative studies exploring the experiences of services users are needed. Finally, interviewees in several of the test bed sites reported that they were still lacking pathways for people with more severe mental health needs who received community sentences, which also needs to be considered in future research. Additional comments and reflections from researchers with lived experience of mental health problems can be found in Box 2.

Conclusions
The experiences of professionals in the test bed sites were largely positive, and stakeholders were motivated to increase use of MHTRs. The MHTR pilot protocol appeared to address a number of previous barriers to the use of MHTRs, including filling a previous gap in services for people with mental health needs who did not meet criteria for secondary-care mental health services, identifying a named clinical lead with regular availability for MHTR assessments, and improving multiagency working and awareness. Having a dedicated team delivering the MHTR intervention also helped the test bed sites to meet the needs of this group of offenders and provide broader support to enable them to engage with the MHTR treatment. Barriers that still need to be addressed include sustainable funding, staffing, motivation for engagement and breach processes. Whether MHTRs provide an appropriate framework for offenders with more severe mental health needs to be further explored. Future work must also evaluate the effectiveness of the new primary care level MHTR pathway in improving both mental health and reoffending.

Funding
Funded through the National Institute for Health Research Policy Research Programme. The views expressed are those of the authors and not necessarily those of the NHS, the National Institute for Health