An out-of-court community-based programme to improve the health and well-being of young adult offenders: the Gateway RCT

Consenting participants

Participation was voluntary. It was not felt appropriate for police investigators to obtain fully informed consent because of the potential risk of coercion, nor was it practical given the timelines. The researchers therefore developed a two-stage consent procedure.

Stage 1 consent

During processing in custody, investigators identified potentially eligible participants (see Appendix 1, Figure 8) and discussed with them the Gateway programme. For legal reasons, the Gateway caution was initially offered as a potential disposal option independently of the study. If interest was shown, the offender was then informed about the ‘Questionnaire Study’ (terminology used in participant facing materials). A Gateway Caution information leaflet (produced by HC independently of the study) and a Questionnaire Study leaflet with link to a video (see Report Supplementary Material 2) were offered and/or e-mailed later. Potential participants were made aware that further details about the study would be provided by a university researcher and that they could withdraw from the study at any time without giving a reason. If the offender agreed to take part in Gateway and the study, the investigator obtained their signature on the combined Stage 1 participant information sheet (PIS) and consent form.

FIGURE 1.

Consolidated Standards of Reporting Trials diagram demonstrating the progression of participants through the trial.

A small number of participants were recruited through the out-of-custody process (see Appendix 1, Figure 9). The investigator followed the process for in-custody, where they were confident that although a disposal decision could not be made at that time, the young person would ultimately be eligible for the Gateway caution option. In these cases, the time between randomisation and disposal could vary. If contacted by telephone, they were asked to give verbal Stage 1 consent to participate. If given, this was recorded in the individual’s Police records management system (RMS) incident record. Written consent was subsequently sought prior to any trial-related activities for the participant. Anyone later declining consent in writing was withdrawn from the trial. This approach ensured all potentially eligible participants had the chance to join the study, in keeping with the pragmatic nature of this trial.

FIGURE 2.

Conceptual framework for determining risk of offending behaviour and role of OOCDs in reducing offending.

Stage 1 consent allowed the police to provide the University of Southampton (UoS) researchers with the participant’s contact details and to access their police records for data on variables such as age, gender and ethnicity and offending history, trigger offence and any subsequent reoffending. Investigators made it clear to potential participants that they would be provided with more details of the study later by the researchers and would have the right to withdraw from the study at any time. The personal contact details of participants who consented to take part in the study were passed to the research team.

Inclusion criteria

• Suspect aged 18–24 years.

• Suspect resided within HC area.

• There was an anticipated guilty plea (i.e. they admitted the offence and said nothing which could be used as a defence or made no admission but had not denied the offence or otherwise indicated it would be contested).

• Full code test met (i.e. there was sufficient evidence to provide a realistic prospect of conviction and it was in the public interest to prosecute or offer a conditional caution to the suspect).

Exclusion criteria

• Hate crime according to Crown Prosecution Service (CPS) Policy.

• Domestic violence-related crime.

• Domestic violence-related crime referred to CPS.

• Sexual offence as defined by the CPS.

• Knife crimes where a decision to prosecute was made.

• Where on conviction the court was more likely to impose a custodial sentence (based on sentencing guides).

• Remand in custody was sought.

• Breach of court or sexual offences orders.

• Any offence involving serious injury or death of another.

• Any serious previous convictions within the last 2 years [i.e. serious violence, grievous bodily harm (GBH) or worse, serious sexual offences, robbery or indictable only offences].

• Summary offences more than 4 months old.

• Persons subject to Court bail, Prison Recall, Red Integrated Offender Management (IOM) or currently under Probation.

• Indictable only offences.

• All drink/drive or endorsable traffic offences.

• Offender had already had a Gateway caution.

• Offender needed an interpreter.

Part 1. Needs assessment

Within 3–5 working days of their disposal, the participant engaged with the Gateway navigator who conducted a thorough needs assessment. Based on identified needs, the navigator assisted the young adult into the appropriate services including with Gateway partner agencies such as alcohol, drug and mental health services. The Gateway navigators were provided by the third sector organisation No Limits and Southampton City Council. The navigators then mentored the individual through the programme. The navigators aimed to hold three face-to-face assessments: 1 at the start, 1 in the middle and the third at the end of the participant’s 16-week caution period. During COVID-19 pandemic restrictions assessments were carried out via phone or video calls. Follow up contact was predominantly via phone, text or video calls.

Part 2. The LINX workshops

The LINX workshops aim to assist young adults in the development of cognitive and affective empathy, accept the need to change attitudes and behaviours including offending and prevent future antisocial and/or violent behaviour. The workshops took place face-to-face in a neutral venue as accessible as possible. Because of COVID-19 social distancing measures, some of the workshops were modified and delivered by telephone or video calls. Details of the differences between the LINX workshops and the LINX telephone delivery are provided in Appendix 2.

LINX workshops for Gateway use carefully constructed experiential group work tools alongside a strong visual framework, ‘Making the LINX to rebuild my life’ wall. LINX workshops seek to enable the young adult to explore and share personal feelings on a variety of issues, particularly around their life experience. The various exercises and activities throughout delivery are designed to take the young adult on a journey, enabling them to see how an experience can create a feeling, which can be translated into a set of behaviours that, for these young adults, may create risk of harm to themselves or others, and thus risk of offending.

Part 3. Condition to not reoffend

If a participant reoffended during the period of their caution, the HC Gateway Team could use their discretion when deciding whether to ‘breach’ the participant (see Appendix 3, Figure 10). If a participant was considered to have breached the terms of the caution, they were withdrawn from the Gateway intervention, and the investigator who originally gave the Gateway Caution considered whether to prosecute the participant for the original offence. Participants who breached their Gateway Conditional Caution continued to be approached for data collection.

Restorative justice

Restorative justice could be added to the standard conditions of the Gateway Caution after discussion and agreement with the young person: the process is victim led, but voluntary for both the victim and the offender. Through RJ, victims can communicate with their offenders and convey the impact the crime has had on them, with the intent to empower the victim. If investigators referred the young person to the RJ process, the navigators and LINX team offered support to the young person, but the actual process was managed as a separate activity. If the victim agreed to RJ conferencing, the young adult would be encouraged to engage with the victim to take positive steps and make amends for the crime committed.

FIGURE 3.

Thematic analysis map – Gateway staff at T1.

Comparator

The comparator was usual process. Under current guidance, for young adults aged 18–24, where there is enough evidence for prosecution (known as Full Code Test) and where the individual admits responsibility, there are various possible outcomes. For less serious offences and where the offender has a limited background of convictions, they may receive a conditional caution. For more serious offences, or where the offender has a more in-depth background in relation to criminal convictions, the offender may be charged and given a court date.

Monitoring adherence to allocations

Spreadsheets within the study case management system, Huddle, provided the Police Gateway Team with oversight of the number of active clients in Gateway, numbers breached, number completed, indicators of breaches/completers, time to date in Gateway and time before breached, discretions applied, monthly recruitment and the numbers refusing participation and cases missed.

An ‘Engagement with client’ spreadsheet was maintained by navigators to record adherence data and was available in Huddle. This spreadsheet included participant ID; type of contact, date of contact, whether participant responded to contact, duration of contact in minutes, name of referring agency and comments from the navigator. Third sector organisations liaised directly with the Police Gateway Team and navigators to report engagement updates in accordance with referrals.

The Hampton Trust provided attendance registers for the LINX workshops/video/calls as these were mandatory sessions for the intervention.

Standard police monitoring of adherence to alternative conditional cautions issued to participants in the usual care group was followed and the information about the conditions and any breaches recorded.

Participant follow up

Participation in the intervention was for 16 weeks from the time of disposal. All participants were asked to take part in telephone interviews at 4-weeks, 16-weeks and 1-year post-randomisation, with flexibility on timing to accommodate the target population. Participants missing one of the data collection points were followed up at the subsequent time point unless they formally withdrew from the study or were withdrawn by the study team during COVID-19 restrictions.

Ahead of each data collection time point, up to four attempts were made to establish contact via text and calls with the participants, with the aim of providing brief information about the study and gauge their availability. If the number was clearly incorrect or out of order, no further contact was attempted. If the researchers were unsuccessful after four attempts, an e-mail was sent to the participant to say that the researchers had been trying to contact them and providing a study phone number for the participant to call. The final contact attempt for non-responders was to send a letter to the participant.

Once an appointment had been booked and the details confirmed with the participant, a text confirmation was sent to their contact mobile number, followed by reminders the day before and the day of the appointment. Discretion was applied to frequency of reminders, guided by interactions with participants.

If the participant cancelled an interview or missed it without notice, the researchers made up to four attempts to re-establish contact and reschedule. If these attempts were unsuccessful, no further attempts were made until the next data collection time point, at which time the process restarted.

The number of contact attempts was indicative, rather than prescriptive. Similarly, flexibility was exercised in relation to the timing of data collection, to accommodate participants’ and researchers’ availability. If no interviews took place at week 4, 16 and 1-year the participant was deemed lost to follow up.

Follow-up data were collected up to 31 March 2022. Attempts to contact participants eligible for week 16 and/or year 1 data collection after the 31 March 2022 were made to inform them that no further involvement was required. A £10 voucher was offered as a gesture of good will.

Outcomes

Adverse event reporting and harms

There were no anticipated adverse events or effects. Potential distress to participants from discussing personal issues or from reflecting on their own behaviour as part of the intervention was monitored by the researchers and the experienced support/case workers delivering the intervention. Referral to appropriate support services could be made where necessary. Participants were informed that if they disclosed any information about a crime the police were unaware of, this would have to be reported to the police.

Participant change of status and withdrawal

Participants were free to withdraw from the study at any point without giving a reason. Each PIS gave information on how a participant could withdraw, including who to contact. Forms for documenting type and reason for withdrawal and other applicable change-of-status categories were available for use by the Police Gateway Team and the UoS researchers.

Participants who consented to Stage 1 but withdrew before giving Stage 2 consent had their study details completely anonymised by the University researchers where personal details had been shared and no study assessments performed.

Participants who consented to Stage 1 but declined Stage 2 consent (without withdrawing) had no study assessments performed.

For participants who withdrew following Stage 2 consent, information already obtained up to that point was retained. To safeguard the individual’s rights under UK general data protection regulation (GDPR) only the minimum personally identifiable information was retained by the Universities. Personal data remained on Huddle, managed by HC, but no longer visible to the researchers and was not downloaded or processed for the purposes of the study. Participant withdrawal after Stage 1 consent required completion of a Change of Status CRF.

Participants who decided to withdraw from the study at any stage did not undergo any further follow-up related to the study.

Loss of capacity during participation in the study

Participants who lost mental capacity after consenting to take part were withdrawn from the study.

Internal pilots

The first 6 months of trial recruitment were designed as an internal pilot to assess feasibility. The study aimed to set up three sites and recruit 50 participants during the pilot. The imposition of a national lockdown in response to the COVID-19 pandemic came at the end of the internal pilot. Progression against the criteria in Table 1 was agreed by the funders and the Study Steering Committee (SSC) and Data Monitoring and Ethics Committee (DMEC). Substantial adaptations to delivery had to be made, so the 6 months after recruitment could safely restart were viewed as a second pilot phase and further progression criteria set (see Table 1).

Sample size

There is no widely accepted and established minimal clinically significant difference for the primary outcome, WEMWBS. It has been suggested that a change of three or more points is likely to be important to individuals, but different statistical approaches provide different estimates ranging from three to eight points (WEMWBS user guide22). There is also variation in the standard deviation (SD) of the WEMWBS with estimates ranging from 6 to 10.828 with the pooled estimate of 10 across all studies. Assuming 90% power, 5% two-sided statistical significance, mean difference of 4 points on WEMWBS and a SD of 10, 266 participants are required. Preliminary figures from The Hampton Trust’s skills/attitudes workshops for domestic abuse (RADAR intervention) suggest a drop-out rate of approximately 15%. Conservatively, assuming a 20% attrition rate, the study aimed to recruit and randomise 334 participants.

Randomisation

Statistical methods

All analyses were conducted in Stata^® version 17 (StataCorp LP, College Station, TX, USA), and reported in accordance with Consolidated Standards of Reporting Trials (CONSORT) guidelines. 29 Version 1.1 of the statistical analysis plan (SAP) was finalised and approved by the SSC/DMEC prior to the completion of data collection on 31 March 2022 (see Report Supplementary Material 4). Version 1.0 of the SAP outlined the planned analyses in order to assess the effectiveness of the Gateway intervention. However, once it was decided by the Trial Management Group (TMG) that due to retention and data collection rates, it was no longer feasible to assess effectiveness. Low retention rates mean a large number of participants would not be included in the primary and secondary analysis models. This is likely to introduce bias into the estimation of the treatment effect, as it is feasible that for this study population to know that whether their data that is missing is dependent on the data values themselves, for example, participants missing WEMWBS data may have lower levels of mental well-being. Version 1.1 of the SAP was therefore produced, removing all reference to formal hypothesis testing and outlining purely descriptive analyses. Continuous measures were summarised using counts, mean, SD, median, IQR, minimum and maximum. Categorical measures were summarised using counts and percentages. All participants were analysed according to their randomised group, unless otherwise stated.

Economic evaluation

Due to the study not reaching its recruitment and data collection targets, it was decided that a health economic analysis was not feasible.

Data collection and management

Public involvement

Three separate approaches were used within the PPI, with the aim of representing all relevant stakeholders to ensure the research was carried out in a way that would be as effective as possible and produce meaningful outputs for those affected. These three areas were with service users (young adult offenders), victims and the public.

Patient and public involvement was embedded early on with the help of partners The Hampton Trust. Meetings with young adults on a Hampton Trust programme explored various aspects of the study, including importance, acceptability and feasibility. The groups fed back in detail around the logistics of the study: the process around consent and randomisation; ways to manage challenges following up the control arm; and opinion on assessment forms.

Once the study was underway, the PPI lead worked with partners to involve young adult representatives who had been through the Gateway programme and those who had been through the ‘usual process’. Consultation and input from these service users provided a clear understanding of the challenges and benefits that participants with and without prior experience of the criminal justice system might face. These PPI representatives worked closely with the PPI lead to develop consent forms, PISs, and initial information leaflets, plan recruitment strategies and consider the most effective ways of arranging interviews and qualitative work.

There were two public representatives on the SSC/DMEC. An ex-offender, working for Hampshire Youth Offenders Team as a peer mentor and support worker; and a victim advocate, working for a charity for victims of crime. They represented the voice of the service users and victims at Steering Group meetings, helping the group reflect on the realities of delivering the programme from the user perspective, reminding the group of some of the vulnerabilities and needs of this population, and ensuring the views of victims were considered.

These two representatives also worked closely with the study PPI lead, providing strategic input, advice and guidance throughout, with a particular focus on the logistics of getting the project underway, reviewing and adapting the protocol. The idea of a recruitment video was conceived by the ex-offender public representative, and the content was co-created with them.

Utilising links established through a local outreach programme, community leaders and members of the public were consulted. The researchers worked closely with these individuals to ensure that they understood the concerns and attitudes of the wider community. Additionally, they were able to provide input to public facing documentation and materials.

Patient and public involvement was able to help the research team consider some of the unique issues facing this vulnerable population. There were challenges, however, and PPI representatives often presented with similar chaotic lives as participants. As a result, involvement was often ad hoc or one-off, with representatives finding it difficult to commit consistently or in a longer term. Working with partners The Hampton Trust and Gateway navigators was invaluable when it came to building trust, identifying and collaborating with offending representatives.

Regulatory approvals and research governance

The outline proposal was submitted to the Hampshire Constabulary Ethics Committee, who agreed to support the study. Any subsequent ethical issues were referred for discussion by HC senior staff.

The study protocol and all associated study documents such as information sheets, consent forms, and questionnaires and subsequent amendments were submitted to the UoS Ethics and Research Governance Board for approval (ERGO Number: 31911).

Confirmation was obtained that the study did not require approvals from the Health Research Authority (HRA), Social Care or Her Majesty Prison Probation Service research ethics committees.

The governance structure for the study comprised the TMG and the independent SSC which also acted as the DMEC. The SSC/DMEC met six times during the course of the study; key meetings were to discuss the set up (October 2018), review of progression and the temporary suspension during COVID-19 restrictions (May 2020), progression following pilot 2 (April 2021) and completion of data collection for early closure of the trial (February 2022). The Chief Investigator had overall responsibility for the study, which was sponsored by the UoS. The TMG submitted regular reports to the funder, SSC/DMEC and sponsor.

Protocol changes and amendments

Protocol amendments were approved by the NIHR Public Health Research (PHR) programme manager, ethics committee and sponsor. A full list of all protocol amendments is provided in Appendix 4 and the rationale for some of the changes explained in Chapter 4.

Trial results

Primary analysis

Table 6 reports the WEMWBS score at 4-weeks, 16-weeks and 1-year post-randomisation. The WEMWBS score can take values between 14 and 70, where higher scores indicate a better state of mental health and well-being. The WEMWBS, SF-12, AUDIT and ADIS data for one participant in the Gateway group was excluded at week 4 due to the questionnaire being completed too early, while at week 16 the data for two participants in the Gateway group were excluded due to the questionnaires being completed too late.

Secondary analyses

Other analyses

Patient public involvement

Young adult offenders represent a vulnerable population with a range of complex and overlapping needs. Recruitment and retention in this cohort is a known challenge, often attributed to these vulnerabilities and chaotic lives. 31,32 For this reason, PPI was considered vital for the success of the study but would come with its own set of unique challenges.

For public involvement to be a meaningful part of the Gateway study, it was essential to invest in strategies to involve this seldom-heard group, rather than PPI members with no real experience of the criminal justice process or the challenges of taking part in research while navigating other lifestyle priorities and vulnerabilities. To address this, the research team took a ‘needs based, place based’ approach to PPI and engagement, respecting the skills and knowledge of partners who have a record of working with this population. The study’s strategy therefore involved user, victim and community engagement.

Summary of trial

The aim of the trial was to evaluate the effectiveness and cost-effectiveness of the Gateway programme, however, the study was unable to obtain sufficient data to undertake any formal hypothesis testing. Similarly, health economic analysis was not appropriate. In the face of consecutive challenges, the researchers have demonstrated that it is possible to recruit and randomise to a RCT in the police setting. The researchers also identified and tested implementation of different approaches to include young people who have committed an offence in research, known to be a difficult group to engage. By working in close collaboration with HC, and their two project dedicated officers, the researchers were able to make pragmatic adjustments to the study design as issues arose. For example, developing a successful approach to recruitment and randomisation by police investigators that included a two-stage consent process. Also, revising the inclusion criteria to encompass all those that HC would consider issuing a Gateway caution to as routine practice. The problems encountered throughout this trial and the team’s endeavours to overcome them are detailed in Chapter 4.

Of those recruited, 191 participants were randomised to the trial (Gateway 109; usual process 82) and included in the descriptive analyses. Although recruitment rates were within the study progression criteria, the number of randomised participants providing data was insufficient to complete the trial (week 4: 49%; week 16: 49%; 1-year: 37%).

Baseline characteristics of the randomised participants were generally well balanced between groups, with small imbalances in gender and highest level of education. Randomised participants and randomised participants who provided primary outcome data at least once, had similar distributions of: age, gender, ethnicity, marital status, level of education, Index of Multiple Deprivation, entry route, RMS incidents involved in, and RMS incidents leading to charge or caution. However, more randomised participants who provided a valid WEMWBS score had a previous conviction compared to the randomised participants.

Participants who attended data collection interviews completed all parts of the WEMWBS, SF-12, AUDIT and ADIS instruments at all time points. Allocation does not appear to have been a significant factor in whether participants attended data collection interviews or not, with similar percentages of those allocated to each arm providing data. Data collection rates show that telephone interviews were acceptable to those willing to share an active telephone number. Rates for those who were non-contactable were similar between the groups at all three time points.

Attempts to contact participants paid off in some cases, but contact dropped off at a similar rate over time in both groups.

Eighty-one of the 105 (77%) participants randomly allocated to the Gateway conditional caution complied with the intervention. Reoffending and non-attendance at the LINX workshops were the reasons for non-compliance.

Although the researchers were able to recruit and randomise 191 participants, the high proportion of participants for whom they did not obtain data at each time point meant there was a high risk of attrition bias. Therefore, an accurate estimate of the effectiveness of the intervention could not be obtained, and as a result statistical hypothesis testing was not carried out.

Sampling, participants and recruitment

An overview of the qualitative evaluation is provided in Table 13.

Due to the small number of participants within the Gateway staff (S) category and potential sensitivity of the data collected, extra care is required to protect identity of the participants. The following citing approaches were used:

• Participants’ job roles were indicated by the ID references, such as N, L or P, next to their quotations. This applied when the role was already suggested by the quote or when the knowledge of the role would be helpful for interpretation of the views expressed.

• When the above considerations do not apply, the Gateway staff will be referred to by the letter S and a number, without an indication of their job role.

Data analysis

Using an inductive approach, reflexive thematic analysis was conducted whereby themes were identified, examined and summarised by an individual coder through an iterative process. The research questions were set before data collection began. However, codes were only created and edited after completion of data collection, during analysis. The themes were formed on the basis of the core ideas that unified participant opinions, which were freely expressed by participants in response to prompts. The type of analysis used was in the tradition of the reflexive thematic analysis, whereby the researcher subjectivity is openly acknowledged. However, the naming of the themes in places tended to be more representative of topic names, rather than acting as an expression of core ideas. This was in part due to the large collection of insights which would not necessarily fit under one idea at the theme level. Data analysis was conducted independently, for each group of participants, at each time period (e.g. clients at T1). Multiple coding was used for a proportion (up to 10%) of the total interviews. As per contemporary approaches to reflexive thematic analysis, the use of multiple coders for a proportion of the data helped check whether interpretations were logical and reasonable, and compare assumptions, rather than achieve a consensus. 37–39 This enabled a more thorough understanding and appreciation of the meaning in the data. The analysis followed the process outlined by Braun and Clarke. 37 An example of how this was applied to the qualitative interviews with young people at T2 is given in Appendix 7.

Time period 1 – Gateway staff

Many of the Gateway staff were already known to members of the research team so individuals were contacted directly. The interviews were conducted by JF, a female professor of criminology and experienced qualitative researcher, AQ, a male post-doctoral researcher in criminology, and SM, a female lecturer in public health. SM and AQ had some experience of qualitative research at the time of interviewing. AQ and SM facilitated the focus groups with the navigators. AQ transcribed the interviews. AQ, SM and IW, a female clinical research fellow in public health, coded the interviews. Several data analysis meetings were held to discuss aspects of the coding.

At the time of interview, all interviewers reported their role at the university and involvement in the study. Given that the researchers in the evaluation team were known to many of the participants, and the evaluation team were academic partners in the project, there would have been some pre-existing beliefs, understanding and ideas around the programme which influenced the types of questions that were asked. All the Gateway staff interviewed at T1, except one HC participant, were female.

The main themes and subthemes are presented in a thematic analysis map in Figure 3. Four key themes were identified: aims and benefits, setup and partnerships, delivery of the intervention and client-related reflections. Quotes from interviews with Gateway staff at T1 that support these themes and subthemes are provided in Report Supplementary Material 5, Table 1.

Subtheme: overall aim of the intervention

Desistance can be defined as abstaining from offending by individuals who previously engaged in such. 40 Desistance was described as the ultimate aim of the Gateway programme, which was understood to provide stability for young people, which would help reduce reoffending. The complex needs of some of the Gateway clients were acknowledged by Gateway staff (FG1-N2).

For high-need individuals, Gateway offered the support where, previously, they often had received very little. The notion of filling a void was prominent, especially in relation to care leavers (N1).

Subtheme: benefits to clients

When asked what benefits Gateway brought to the service users, Gateway staff described their personal experiences working with clients. Despite it being ‘early days’ at T1, they described beneficial results, with some exceptions. For example, with clients for whom this type of support was insufficient to make a difference (N1).

In terms of the benefits that were directly, and immediately, experienced by the clients, the staff felt that Gateway answered the otherwise unfulfilled young people’s need to have someone to talk to – someone who would listen without judgment, including on topics that they would not be able to talk about to anyone else. This open conversation started from the time of the initial assessment (FG1-N1). Young people had reported to the staff the feeling of being believed in, which was triggered by the support within Gateway and was instrumental in building their self-esteem (L2-2). They also provided feedback that they were more comfortable around new people following completion of the LINX workshops. It was taken as a sign that they would be more likely to ask for help, which, in the eyes of the staff, was a positive development (L2-3). Another benefit was that of being dealt with out-of-court and without prosecution; not receiving a criminal conviction mattered (FG1-N4).

The navigators were able to offer practical support to enable the clients to pursue employment or educational opportunities, or secure housing. It was humbling to hear that some young people did not have a form of identification prior to Gateway, which precluded them from accessing housing, benefits or job opportunities, and how truly life-changing receiving an identity card with the help of a navigator was for them (P1-2).

In comparison to other disposals offered to young people by the police, Gateway gave clients the opportunity to get support to resolve underlying issues, offering a tailored approach, for example, support with overcoming drug addiction if that was felt to be at the core of their offending (L1-1).

In summary, this theme revealed that the staff viewed the Gateway programme as aimed at improving lives and reducing, even if not necessarily stopping, reoffending. It was seen as benefiting young people who would otherwise go through alternative pathways that would address their underlying needs and, in some cases, could lead to a conviction, impacting their life chances further. Gateway aimed to fill a service gap, offering both an opportunity to forge trusting relationships with staff and receive advice and practical support, which were vital, in particular, to those with complex needs.

Subtheme: multiagency partnerships

Gateway was a multiagency intervention. Navigators also worked with, or referred their clients to, other organisations outside of the programme. Some young people were engaged with other support agencies when they came to Gateway. In such cases their navigator aimed to work closely with other support workers, for example, from the community mental health team (N1).

The Gateway programme was being delivered by three separate types of organisations (the police, two navigator providers and the LINX provider). Working with other agencies was considered to be a challenge, which could, in turn, affect the relationships between the navigators and young people. Time and again, wishes were expressed for better partnership working, by staff from every stakeholder organisation. Related issues affected both the staff, and indirectly, likely also the young people. Working across agencies was seen as challenging, especially due to communication difficulties, including keeping all parties updated. Different line management structures, as well as priorities, policies and procedures were cited as some of the factors contributing to the discord (S4, S3, S1).

The challenges in communication (defined in the next sub-theme) were also compounded by staff turnover, making it difficult for staff to have a shared, up-to-date understanding of the aims and deliverables of the Gateway programme, even within an individual agency. Some of the staff had already had three managers by their interview in this time period, and it felt disruptive having to educate the supervisors about the intervention, especially considering the complexity and the challenging nature of Gateway (S3).

The navigators, who supported the clients on the Gateway programme, were primarily based in the office space of the lead organisation, HC. As a result of this, they felt estranged from the navigator provider organisations, which employed them, since they spent most of their desk time at Southampton Central Police Station. It took a very long time for the staff at one of the provider organisations to realise that the navigators were also employees there, due to the lack of daily contact; similarly, there were long delays in receiving access to provider organisation e-mail (S7).

Subtheme: navigator location

At T1, before the COVID-19 pandemic, the navigators shared physical office space with the Gateway police officers; an arrangement that had its benefits and drawbacks. There were difficulties associated with the fact that the navigators were supposed to discuss certain issues concerning the clients with their navigator provider managers, who may be less informed on a specific issue, rather than with the police officers with whom the navigators shared office space. This was due to the duty of the police officers to report information as intelligence (S7). The duty of the police to report certain information about, or obtained from, the Gateway clients as intelligence, was a somewhat unforeseen, but significant, issue. Approaches to confidentiality and what needs to be shared between partners or reported to the police also differed between organisations, which created friction (S1).

Subtheme: communication and information-sharing

Challenges associated with sharing of information were frequently raised by interviewees from all the partner organisations. When Gateway was set up, all partners (the police, two navigator providers and LINX provider) had access to a shared IT system, so they could access information about Gateway clients. All staff were also routinely involved in reporting and sharing information about clients, including that which emerged in the course of LINX workshops and the provision of individualised support by their navigator. This enabled identification of further needs and further tailoring of support (S3).

However, over time, due to differences in policies and processes around sharing confidential information, this practice stopped. Sharing between the partners became limited to information relating to safeguarding and attendance at sessions by clients (S11). Information was not routinely shared between the navigators and LINX practitioners, unless consent was explicitly gained from the client. This was challenging, as it was felt that opportunities to put extra, or more tailored, support in place were missed (S7, L2-2).

This meant that at T1, there was no feedback loop in relation to information shared between practitioners from different organisations, which made it challenging to work cohesively together. Nonetheless, the partners within the Gateway programme recognised the need for efficient, collaborative communication, while being all too aware of some deficiencies (S10).

Subtheme: LINX workshops

The 2-day LINX workshop was an integral part of the Gateway programme. It was created by The Hampton Trust and underwent continuous development thereafter. It was designed to be thought-provoking and encouraging self-reflection and examination of the clients’ own past, present and future, including any factors which may have contributed to their offending. This gave the LINK facilitators a chance to discuss with the clients how they might want to work through issues with their navigators (L1-1). Clients shared with their navigators the ways in which the workshops allowed them to change previously unhelpful reactions to events (P1-2).

Subtheme: navigators

This subtheme covers any aspects relating to the routine work of the navigators. In a typical working day, a navigator may complete initial, midway and final assessments for clients, continue earlier work with existing clients and deal with any issues requiring attention (N2).

At T1, a very prominent topic, brought up in most of the interviews and focus group discussions, was that of the low numbers of Gateway clients. The navigators identified pros and cons to this. It was noted that the numbers coming through were generally very small, albeit the numbers fluctuated from month to month (N3, N2).

It was reported that some navigators had to take on other duties with their employing organisation, due to the low intake of Gateway clients, which at the time of the interview was insufficient to occupy the four navigators in post, even with the pre-pandemic ways of working, that is with a heavy face-to-face involvement (N2).

Subtheme: eligibility and recruitment

The issue of the low numbers of young people eligible for recruitment has been present from the beginning of implementation of the intervention, and various steps have been taken to alleviate this. The intervention was launched in June 2018, and was altered in due course, with Gateway becoming a type of conditional caution, rather just simply an alternative to prosecution or court appearance. This was intended to boost numbers of eligible people and led to more low-end-of-offending clients being recruited, which still felt would be beneficial for some clients (P2-1, FG3-N2).

Subtheme: experience and training

On the whole, the navigators felt well prepared for their roles, while acknowledging that there were some gaps in their training, and that they had unequal access to different types of training, depending on which organisation employed them. This was seen as a challenge, since all the navigators required similar training, it was felt. Some of the training received through one of the navigator provider organisations was aimed at staff working with younger people, and was seen as less relevant (N1, N3). Importance of mental health training was often emphasised, and the navigators also relied on previous training received elsewhere (N1, FG1-N2).

General training about services available may have been helpful for LINX practitioners, had it been offered, albeit this may be more directly relevant to role of the navigators (L2-2).

Subtheme: job satisfaction

Despite the challenges cited by the participants, it was clear that they also derived great satisfaction from their jobs. They enjoyed case-working and took pride in getting positive results with the clients (N1). This was found to be motivational when challenges arose (L2-2). Aspects such as opportunities to demonstrate innovation and find a specific approach to help individual clients were much appreciated and helped them remain passionate about their work (P2-2).

Subtheme: needs

The participants frequently referred to the wide spectrum of needs clients presented with. This subtheme often overlapped with that of client backgrounds. Clients really varied in the degree of support they needed: from people with very complex backgrounds with much adversity, to those who made one mistake bringing them into contact with the criminal justice system. This meant a degree of flexibility was required in approach to the different types of clients (L2-2, FG1-N2). Women’s needs could differ from men’s, and the workshops, for example, were more typically suited to men (L2-2).

Subtheme: attitude towards Gateway

This subtheme was concerned with clients’ motivation and intrinsic belief in the value of Gateway. Much of this the staff found out from the written feedback at the end of the programme. The ability to share and be listened to, without prejudice, as mentioned, which was confidence boosting (L2-2). The young people seemed to enjoy the intervention, although it was presupposed that it may have felt like too much work for some (N1). Readiness for change was seen as key, but it was a difficult task to help convince some young people of the need to change (SM3). Occasionally, young people remained closed to the idea of help, at least in the shape it was being provided, maintaining that some people won’t need help and denying deriving any benefit from Gateway, sometimes quite forcefully (L2-2).

Subtheme: engagement

The staff in all categories mentioned the importance of building a rapport with the young people as important to ensuring their engagement with the intervention. Building a rapport (while bearing in mind any risks that the young person may present) was seen as crucial and an important part of the job (FG1-N1).

The navigators reported that engagement was variable. It was cited that a contributing factor may have been a lack of appreciation during the recruitment process of how much involvement was required. Non-engagement constituted one of the reasons for which Gateway clients could be breached, but there had been discrepancies in interpretation of how low engagement would need to be in order to constitute a breech (FG3-N2). There were at times differences in opinions with regards to individual breaching decisions, and the project officers making these appreciated the complexity and treated breeching decisions as discretionary, requiring individual consideration (P2-2).

Subtheme: social determinants and their impacts

Participants discussed a number of social determinants and their impacts, often identifying these factors as the cause of offending (P010, P004). Individuals who perceived these needs acutely were more likely to feel the navigator sessions had an impact on their lives (P017). Participants without a high perception of need, or who did not view the crime they committed as problematic, were less likely to consider sessions valuable (P005). The limitations of Gateway to support social determinants at a societal level was reflected upon (P010).

Subtheme: LINX workshops

Participants felt anxious or apprehensive of group work beforehand (P015, P004, P011), then discussed enjoying the sessions and feeling put at ease (P006). The workshops being fun stood out over content for some while others felt that the workshops were not relevant to them (P017). A key reflection was that of a change in perspective (P014), particularly in relation to seeing different perspectives (P005). Seeing things from another’s point of view and discussing other’s crimes helped participants to feel less alone in relation to their offences (P015) appearing to alleviate some of the isolation experienced going through the process (P017), in turn raising self-esteem (P015).

Subtheme: barriers, facilitators and engagement

Participants had negative expectations (P006), particularly if they believed Gateway would be undertaken by the police (P004). Despite this, the participants interviewed felt that they engaged well with the programme; attributing this to the rapport built with staff (P011). One participant who breached for lack of engagement perceived the programme to be easy to engage with (P005). Logistically, most participants felt that there were no real barriers to attending navigator sessions, reflecting positively on the flexibility of staff (P007, P017). Those with competing priorities, such as work or childcare, reported some challenges working around commitments but felt expectations were fair (P004). Attending LINX workshops was felt to be more challenging, largely due to the lack of flexibility in time and location, and the length (P004).

Subtheme: motivation

Participants’ initial motivation was to avoid going to court (P005), with little emphasis placed on the purpose of the programme in terms of health and well-being (P006). However, motivations changed throughout the Gateway programme, with some reflecting on the shift from fear of consequences, to no longer wanting to offend (P007). As participants built a rapport with their navigator, the idea of not wanting to let them down also became key (P005, P004).

Subtheme: relationships, and relationship with navigator

Relationships with families, and the support they provided impacted help-seeking behaviour prior to Gateway. Where participants had supportive and positive parental role models, they were not equipped to provide adequate support (P017, P018). Parents also played a key role in participants engaging; for several, family facilitated their attendance (P015). The relationship that participants developed with their navigator was a crucial aspect that contributed towards the perceived benefits (P015). Navigators were widely reported as going ‘over and above’ to provide personalised support for participants (P017, P004), and participants perceived them to be genuinely well meaning and authentic (P015).

Subtheme: control and ownership

Participants discussed feeling out of control of their lives before being arrested (P017), with Gateway being initially seen as an extension of this (P011). Participants viewed Gateway as a choice, and this perception of self-determination evoked positive attitudes (P017). Individuals who felt they had ownership over the programme were more likely to see benefit (P004), with some reflecting on the wider impact that control gave over their sense of self-esteem (P017). However, external factors contributed to some participants, sense of control to stop reoffending. The participant who breached for reoffending (carrying a knife) described not having control over his offending behaviour (P001).

Subtheme: hope, aspiration and optimism

Participants viewed Gateway as a second chance, either as an opportunity not to be criminalised, to make a change and/or learn from an offence (P005, P010, P007). Within the programme, the sense of being believed in was tied in with the relationship participants had with their navigator (P017), with participants reflecting on the feeling that someone understood them for the first time (P014). The description of being ‘fixed’ by the navigators, highlighted the perception of feeling ‘broken’ prior to the change (P017). Participants described having hope and optimism for the future, particularly if they felt they were in a position that they had some control over its direction (P015).

Subtheme: success means different things to different people

Perceptions of success varied widely. When recidivism was discussed, it was clear that for many participants success was about offending less, rather than not offending (P004). For some, addressing the social determinants, or their impacts was a tangible and meaningful measure of success (P012, P001). Others described positive changes to their mental health as being the most significant achievement. Learning about the drivers of offending behaviour helped participants to understand their own offending, recognise how Gateway had helped them build trust and seek help where they would not have before, and reflected on how they felt more prepared to manage situations (P010, P014, P015). Several participants discussed the knock-on effect of the programme in their wider lives, describing Gateway as a catalyst for change (P010). Participants who were parents described the positive impact that the programme had on their children (P014).

Subtheme: losing support after the programme

Participants reflected on the impact of the programme ending and support being lost. For example, access to the council housing register facilitated by a navigator was lost, indicating how easily progress made can be lost (P010). Participants identified the short-term impact from the programme but described how things worsened over time (P010, P011). Feelings were expressed that, despite the programme’s support, it was not enough to counter the wider environmental influences long-term (P001). Requests to keep in touch with navigators led to the positive impact of having someone there to celebrate successes (P017, P004).

Demographics of young people interviewed

Table 14 presents the demographics of the young people interviewed. Participants were predominantly male. All had some experience of the navigator delivering their component by phone. Nine participants had in-person LINX workshops, six were by phone and one was unspecified. To protect anonymity, the details regarding HC site locations have been removed, however, the participants represent a total of 12 sites across Hampshire. Most participants had also participated in the Questionnaire Study (RTC) component. The research interviews were conducted by LW, a female interviewer working on the study.

While the discussion guide was designed to allow for the exploration of eight themes, the analysis of the interview data revealed three broad themes and twelve subthemes (see Figure 5). These themes represent the young people’s views on (1) barriers and enablers for change, (2) impressions of Gateway and (3) reported benefits. They are discussed below, along with their relevant subthemes. Quotes from interviews with Gateway clients at T3 are available in Report Supplementary Material 5, Table 3.

Subtheme: relationship with the navigator

When discussing their relationship with the navigator, the overall response of the young people was very positive. In all their accounts the participants reported feeling supported, and there was a unanimous view that the navigators had been friendly, understanding and approachable. Frequent words used to describe their navigators included nice, lovely, sweet, genuine, helpful and even perfect (C7).

It was felt that such characteristics relieved initial feelings of apprehension and uncertainty while also instilling trust and confidence in the programme’s confidentiality. Many of the participants mentioned this was helpful with engagement. Approachability, openness and calmness in communication were highly valued and encouraged disclosure (C7, C9, C11).

Contrary to initial expectations, the young people described their experience with their navigator as being non-judgmental, and a sense of ease was expressed in relation to their understanding and accepting nature. The sessions provided a safe environment in which emotions brought on by difficult or frustrating topics could be better maintained.

Except for one participant, none of the young people had negative comments regarding this aspect of the programme. Furthermore, participants reflected on previous difficulties and experiences with past care workers, suggesting it was rare to have developed such a relationship with a professional of this kind (C11).

It was widely acknowledged that the navigators’ flexibility accommodated a variety of schedules and pre-existing commitments. This was considered valuable in preventing logistical barriers and thus allowed for greater success with compliance. Participants also demonstrated gratitude for the navigators’ reliability in terms of timekeeping and prompt responses to e-mails and text messages (C13). A further commonality among the participants was their recognition of the navigators’ desire to make a difference and the navigators coming across as thoroughly enjoying their jobs (C9).

It was felt that the provision of practical support and signposting facilitated better opportunities in terms of employment, finances and housing. In addition to this, the young people reported the introduction to, and consequent utilisation of additional services such as food banks.

The overall responses regarding the navigators demonstrate the significance of their role in enabling and implementing change. Their ability to gain trust through a friendly and approachable demeanour appeared valuable in instilling confidence in the participants, allowing for greater engagement, and thus, potential to benefit from the experience.

Subtheme: LINX

When questioned about the LINX component, the young people largely demonstrated recognition of its ability to enable better decision-making and coping skills. The interviewees cited acquiring the necessary tools to manage and deflect challenging emotions such as stress and anger, as well as strategies to help control drinking. Some reported having used these tools since completing the programme (C13).

When speaking of the staff, the participants used descriptive words similar to those associated with the navigators. Examples included friendly, understanding, lovely and non-judgmental. Participants were being treated in a non-critical, respectful manner, without being associated with their offence. It was felt that such characteristics of the staff were, again, valuable in improving the young people’s confidence when engaging (C13, C8).

A range of responses were elicited from questions regarding the young people’s feelings towards participating and sharing in a group environment. It was felt this offered a valuable opportunity to engage with individuals in similar situations. This appeared to boost morale as participants reflected on joining together in a non-judgmental environment. Participants were often surprised that other young people in the workshops were in general friendly, and the atmosphere was relaxed (C13). Clients made connections with others in similar situations (C8). Furthermore, the benefit of considering a variety of perspectives was believed to have enabled clearer decision-making.

Contrary to this, there were also reports of a sense of isolation from the group due to demographic factors and a difference in offence type, and hence feeling ostracised (C3). This highlighted a potential barrier in terms of engagement and compliance.

Typically, participants found the prospect of group work intimidating and described initial feelings of apprehension and awkwardness. This component could be seen as unhelpful due to the challenges involved with participating in such a setting. The perceived expectation of being grouped with fellow ‘criminals’ was also considered to have initiated anxiety (C13).

It was felt by some that such apprehension and uncertainty affected attendance, which was reportedly low. Some felt this could be due to insufficient information at the start of the programme leading to poor, or uninformed expectations. This is defined in the later subtheme, initial communication. A minority also advised that attendance depended on the ability to take the time off work. While inconvenient, none of the participants reported this to have been problematic.

Issues relating to the location were particularly prevalent. While a substantial minority commented on its convenience, most participants reported long travel times and accompanying costs. Some reflected with a sense of confusion due to the perception that it was inconvenient for all, including the course leaders. Except for one participant, all respondents reported requiring a family member, taxi, or public transport to attend. It was further reported that the venues had been inaccessible by public transport and therefore, in addition to a train journey, then required a substantial walk either side to attend. This was felt to greatly contribute to the duration of the day. Separately, the duration was considered an additional challenge with participants advising it to be too long to maintain concentration (C13, C9).

Participants also commented on the associated travel costs, which could be considerable, depending on the location (C9, C13). It was suggested that similar group sizes should be held in different locations, making it more accessible and convenient for attendees. Further comments also included the suggestion that participants be fully or partly reimbursed for travel.

A switch from face-to-face workshops to phone delivery during COVID-19 appeared to have varying effects. It was common that participants had difficulties differentiating these sessions from those with their navigator, and thus many struggled to remember the LINX component. In general, however, participants stated that the phone delivery offered some benefits. In addition to greater convenience and thus compliance, it was felt that better confidence could be achieved with engagement. This is defined further in the later subtheme, societal issues and COVID-19.

Subtheme: tailoring

This subtheme emerged as participants reflected on the programme’s ability to accommodate varying needs. A variety of perspectives were expressed regarding how the LINX workshops should be organised, as well as how participants felt their individual needs were met.

Conflicting views emerged when speaking of group work. There were views that providing for a variety of needs allowed for a more diverse learning experience. Here they reported to have received multiple life lessons and strategies to deal with a variety of circumstances. Perhaps surprising to note is this view was largely expressed by participants seemingly of lower need status. Such responses suggested they had found the workshops informative and interesting (C13).

Others felt the content of the group work was not relevant to themselves and thus, offered little benefit to their situation. For example, a large proportion of the workshop had been focused on anger management despite some people’s offences having no association with anger. In addition, the workshops were also reported to be of a generic nature and aimed at a younger audience, with reports of not having learned anything from the experience (C9).

Participants also demonstrated confusion over the issue of people with different offences taking part in the workshops together, questioning the reasons behind grouping in this manner. Furthermore, a minority felt that it was harder to comprehend and follow the content of the workshops due to quick changes in topics and variety of material. It was therefore suggested that better personalisation could be achieved by grouping according to offence type or level of need (C3).

Speaking of their experience with the navigator, the clients felt that the level of support was appropriate for their level of need. Participants commented on their ability to understand individual circumstances, pick up on cues from the participants and tailor their delivery accordingly (C13, C4). It was reported that there was an appreciation that not everyone needed intense support (C14).

This was in contrast to the view that greater emphasis could have been placed on the topic of mental health. They commented, ‘It just wasn’t tailored for me’, adding later, ‘it was just preaching the wrong stuff to me, personally’ and, ‘I just didn’t find it helpful’ (C9).

Subtheme: talking, trust and engagement

When asked whether there were any aspects of the programme found to be particularly helpful, many of the participants commented on the therapeutic opportunity of talking. Specifically noted was the benefit of talking with anonymity. In their accounts, participants reported feeling comforted by having someone to listen to them and to have their voices heard. The opportunity to explain to someone how they felt and to describe their point of view was frequently described as a sense of weight being lifted (C4).

Talking was also considered an enabling factor in achieving greater insight. Participants spoke of the opportunity to reflect on their lives and suggested that this had allowed for the realisation of the need to initiate change. For some, this was recognising the necessity of reducing drug or alcohol use, while for others it was leaving an unhealthy relationship or changing their mindset. Other participants felt that it allowed for greater clarity in decision-making, while further, some advised that it provided an opportunity to obtain additional help, for example, in order to change their lifestyle (C1, C11).

Participants reported challenges with engaging in sensitive and uncomfortable topics. It was felt that this therefore had the potential to hinder conversations. Furthermore, contrary to the above, sometimes there was a reluctance to engage with a stranger, especially on painful topics (C13, C7).

Negative experiences with past care workers had the potential to lead to poor expectations and consequently presented a further barrier to engagement. There were reports of feeling let down by previous counsellors, citing feelings of abandonment and frustration relating to high staff turnover rates. It was felt that this had a detrimental effect on their ability to trust the staff, as well as their willingness to continue seeking help (C4).

An additional barrier was presented by the participants’ ability to engage. Non-engagement of LINX was the cause of one participant’s breach. While this individual also failed to engage in the study interview, another openly disclosed their struggles with alcohol dependency. Speaking of the LINX workshops, they confided that they attended while under the influence of alcohol. This was also the case when participating in the study interview. Such alcohol dependency was, for this participant, the motivating factor for accepting Gateway (C3).

Comments regarding trust were also prevalent. Through their accounts, the participants presented a variety of perspectives, and thus it was considered to present as both a barrier, and an enabler for change.

It was widely acknowledged that a significant concern was that of the programme’s association with the police. From their accounts, it was clear that the participants’ views on the police were very negative, and thus introduced an initial barrier. Talking about this issue, one interviewee alluded that such association had the potential to impact engagement and lead to poor expectations. It is therefore possible to suggest that some young people may decline Gateway due to this association (C7).

It was reported that an initial lack of trust could prevent participants from fully engaging with their navigator at the start of the programme. This may have been through shutting down due to a fear that talking could have detrimental consequences in the form of further punishment from the police (C4).

Contrary to such opinions, there were also expressions of confidence in the programme’s confidentiality and reports of this positively influencing engagement (C10).

The result of losing trust was, however, incredibly detrimental, illustrating the necessity of transparent communication. In one case, a misunderstanding of their navigator’s identity led to feelings of betrayal, withdrawal and paranoia (C3).

The need to differentiate Gateway staff from the police therefore appears imperative in optimising compliance, engagement and success.

Subtheme: societal issues and COVID-19

This subtheme occurred as participants spoke of the barriers presented on a societal level. It was felt that there is very limited help available outside of Gateway, particularly for this age range. One participant alluded to a sense of abandonment as they reported being no longer able to work with them since turning 25. There were reports of no help being available, while others cited long waiting lists and associated costs with private care. It was also apparent that the young people were unaware of what help was available, or how to access it (C9).

In the absence of available support, it was felt by one participant that affordability may have presented an important barrier to change. While unemployed, it is difficult to find work to finance a rehabilitation service for alcohol dependency (C3).

Some participants volunteered that their mental health had worsened following the COVID-19 pandemic. This was attributed to the difficulties it presented in terms of employment, living circumstances and lack of leisure opportunities. Furthermore, one participant reported it to be a contributing factor to their offence. Despite this, except for a reported delay in the processing of paperwork, issues concerning the effect of the pandemic on the intervention were not prominent in the interview data.

While a minority felt that their experience may have been more personal if it had been face-to-face, the majority agreed that the phone intervention provided a convenient alternative and had little impact on engagement or their ability to benefit. In addition to this, participants felt better able to open up on the phone, suggesting it to be a less daunting option and thus enabled greater confidence when engaging (C4).

Subtheme: expectations

When questioned about their expectations of the programme, most of the young people’s responses were negative. Many described expecting it to be lecture based and that they would be spoken to about their mistakes and given lessons on how to behave. It was widely anticipated that the staff would be judgmental and not allow participants to have their own say (C8).

Such negative expectations could be somewhat attributed to the association with the police. Participants cited expecting to be questioned, and that due to the location at the police station, the programme would therefore be presented by the police themselves (C1, C7). Contrary to this, a minority of the young people also spoke of positive expectations. They cited expecting the programme to help them in their lives by providing the necessary support needed to instigate lifestyle changes as well as the opportunity to talk (C10).

Reports of anxiety, doubt and scepticism were widespread. Participants spoke of initial uncertainty and doubt as to whether Gateway would help their situation or provide any benefit. Furthermore, one participant reported their dislike of classrooms was off-putting to the extent they considered taking standard prosecution.

A large proportion of participants reported not having known what to expect. Some suggested this was due to a lack of information when being offered the programme. This is described in the following subtheme, initial communication.

Subtheme: initial communication

A recurrent theme in the interviews was the opinion that there was a need for greater initial information and better communication when offering the programme to prospective clients.

Participants indicated that they had received adequate information and were therefore clear on what it involved. Some voiced concerns that a lack of initial information could result in young people accepting the caution without fully understanding what to expect, with calls for better information. Some went on to suggest that discovering this later could impact compliance, resulting in breaches and thus worse consequences than accepting standard prosecution (C13). This had the potential to affect engagement due to negative expectations (C4).

While expressing gratitude for the help received there was disappointment over other areas not meeting expectations. It was felt that the cause was an inability to fully comprehend the information to begin with. An example is inability to understand the information due to it being presented in writing only. Such a lack of understanding could potentially lead to uninformed consent, incorrect expectations and disappointment (C3).

Subtheme: attitude

While such concerns were reported regarding initial communication, ease of location and tailoring, most participants reflected on the programme with a sense of positivity and appreciation (C4, C5, C3).

The participants demonstrated appreciation for Gateway providing the opportunity of chance. The young people recognised it as an opportunity to rehabilitate and rectify certain elements of their lives. Participants commented on its additional value for those younger than themselves (C12).

Contrary to this, there was also a demonstration of an embedded lack of acceptance, questioning the reasons behind being cautioned for a mental health-related offence and reporting a feeling that they had accepted Gateway for the wrong reasons, and that there was nothing they perceived themselves as needing help with. As a result, there were no benefits to report.

Positive views were often portrayed as transitions of attitudes. For example, initial feelings of apprehension and scepticism were reportedly replaced by a sense of acceptance and acknowledgement of the programme’s ability to help (C7, C4). In general, participants described their experience as being better than anticipated (C12).

When questioned about their motivations for participating in the research study, many of the responses further demonstrated the young people’s appreciation for the programme. While it was clear that the use of incentives was successful in optimising compliance, many participants reported a willingness to help, regardless of the reward. The desire to give back and help others by participating was felt to be driven by their own personal results and successes which they attributed to the programme.

Subtheme: managing self

Young people reported having learned better coping strategies. An improvement in the ability to manage emotions was associated with a reported positive effect on behaviour. Participants reported being able to defuse situations to prevent arguments, as well as redirect challenging emotions into positive activities such as the gym or reading. This was felt to improve relationships (C4).

One participant also stated that since finishing the programme they had started to distance themselves from former friends known to be associated with risky behaviour such as substance use (C8).

Questions about recidivism brought forth a variety of responses. The majority reported they would not reoffend, however not all attributed this to Gateway. Many of the participants considered their offence to be a one-off mistake, while others advised that being arrested was the wake-up call needed to prompt change. One participant demonstrated no remorse for their offence, advising instead that they would continue the activity due to a lifetime of knowing no different. Responses also included those crediting the programme (C8, C10, C7).

Subtheme: personal circumstances

This subtheme emerged as the young people discussed positive changes in terms of employment, housing and relationships. A change in mindset was associated with a greater desire to seek meaningful work as well as a new-found commitment to staying employed (C8).

In addition to improved efficiency in everyday life, an ability to work longer hours was also reported (C4).

When asked about any changes they may have experienced in their housing situation, many of the young people felt they had achieved greater stability. Some revealed they were no longer homeless, while others reported having moved from hostels or their family homes. In some circumstances the young people’s living situations were considered a risk factor in relation to offending. The main cause cited for this was tensions within families. It was therefore felt that moving from that family environment resulted in improved relationships, reduced arguments and a perceived lower risk of reoffending (C13).

While many participants felt that their relationships had improved, a minority cited continuous difficulties due to a lack of family support, acceptance of their caution, or forgiveness for their offence (C16).

Regardless of employment status, several participants considered applying for Jobseekers Allowance to be a benefit of completing the programme. Many demonstrated gratitude for their navigators facilitating opportunities through their help with the various steps involved. Such challenges included difficulties with computer literacy and therefore highlighted pre-existing barriers in terms of accessibility for these young people.

Improved employment chances combined with the provision of financial advice and support gave greater opportunity for further stability. Some participants reflected on this improvement in their lives with a sense of disbelief (C7).

Subtheme: health

The participants, to varying extents, spoke of improvements in their health. Reports of reduced drug and alcohol use were particularly prevalent. Some participants acknowledged this benefited not only their health, but also suggested it to be a contributing factor in their reduced likelihood to reoffend.

Comments regarding mental health also recurred throughout the interviews. Participants spoke of the programme’s ability to reinforce their positive outlook and improve their mental well-being as a result (C7, C8, C11).

Participants reported that the programme helped establish routine in their lives, helping them stay better organised on a daily basis (C3).

There was also a reported increase in the utilisation of services, with participants advising they had returned to their general practitioners and dentists (C3). Better management of prescriptions was associated with improved employment opportunities (C1).

The programme was also credited for helping the young people’s physical health. Participants reported taking better care of themselves, hiring personal trainers, attending the gym and participating in healthy leisure activities such as walking. This was also cited as a coping strategy to better manage challenging emotions. The study’s voucher incentive, for example, could be used to buy gym clothes in order to support a new lifestyle (C7).

Subtheme: no benefits

A minority of the young people demonstrated resistance towards the programme. Some did not perceive their offences as being troublesome while others felt that those relating to mental health should not result in a caution. It was also felt by one participant that the programme was perhaps better suited to those on their second or third offence. This introduced a question over the participants’ suitability for Gateway. As mentioned previously, while reporting to have found the programme interesting, some of the young people cited having well-paid jobs, supportive families, and a good education. Several individuals felt that they had just made a one-off mistake which was acknowledged at the time. Their opinions on benefits varied. Some felt they had gained little or no benefit from taking part, yet recognised that due to the pre-existing stability of their lives, there was little Gateway could offer them. Others felt they benefited from gaining insight, recognising certain aspects needed to change. Avoiding prosecution was also considered a benefit (C14).

While previously suggesting cognitive behavioural therapy may be helpful in their situation, another participant also felt that there was nothing they needed help with (C9).

Subtheme: training

This subtheme covered the training received both before and during working on Gateway, as well as any previous experience which the staff brought into their role. Most participants reported receiving substantial training both prior to commencing in the post and since. Understandably, the training requirements differed between the three categories of the staff, with the highest requirements in the Gateway navigators, who worked very closely with the clients and helped them address a wide variety of needs. The navigators in particular reported receiving extensive training while in post (N2).

As previously mentioned, some of the training was directed at helping children and young people younger than 18 years, but overall, it was felt that the skills were transferrable. Gathering information with regards to local services available for 18- to 24-year-olds was seen as part of the job in any case, so the focus of some of the training on the younger population was not felt to be an issue (N1).

Subtheme: limitations of Gateway

The subtheme of limitations of the programme emerged as the staff reflected on the challenges they have encountered while working on Gateway. The need for a tailored approach to clients was a particularly prominent topic. It was acknowledged that Gateway was not for everyone, and that for some clients, perhaps, a simple caution or, for example, a drug awareness course would have been more appropriate (S2, S3). On the other hand, it was discussed that the higher-need clients often needed a longer period of time on the programme than the standard 16-weeks.

The staff also commented on the fact that young people under the highly stressful circumstances of being questioned by the police may not process the information fully and give consent without realising the involvement required (S7).

Subtheme: LINX workshops

The LINX part of the intervention consisted of two in-person group workshops, or telephone calls in the place of the workshops, when this was dictated by the COVID-19 restrictions or when there were insufficient client numbers for groups. LINX workshops were seen as an integral part of the intervention, which was at the same time somewhat separate from the navigator support. LINX workshops were reported to act as something of a trigger, an opportunity to make connections between events in their own lives, or to bring up the topic of mental health and its importance (HT2).

The groups were arranged according to gender, but Included young people of varying levels of need, some could be university students or in full-time employment and/or from very supportive families, all the way to people who had very traumatic childhoods and had very high levels of need. It was reported that this could create a certain divide in some groups, whereby the young people were not able to relate to each other (HT1). Yet, this fact was seen as potentially beneficial, whereby clients from more stable circumstances could view these encounters as a wake-up call, and for some other young people meeting those in better circumstances could be inspiring (HT2, HT3). Occasionally, however, a whole group could have consisted of low-need participants, and the nature of the group work as originally planned would be seen as largely irrelevant to them (HT1).

As reported above, the navigators saw tailoring the intervention to the level of need as key, including as a way to ensure that the low-need clients are not overburdened with the input they do not need and which may be disrupting their lives. On the other hand, the LINX facilitators, who only interacted with the clients on two occasions, did not see the involvement of the low-need clients as a wholly negative phenomenon. While acknowledging the gap that existed between some of the young people in terms of their background, some of the young people may have felt inspired (HT2, HT3), potentially encouraging them to step out of their comfort zone and, for example, apply for jobs. However, it was acknowledged that groups with clients from a mixture of backgrounds were more difficult to manage (HT1, HT2, HT3).

Subtheme: COVID-19

The effect of the COVID-19 pandemic was explored with all the staff participants. There was a wide range of opinions on the effect of the pandemic on the work within the Gateway project, on the lives of the Gateway clients and on their compliance with the intervention. Several interviewees volunteered that there had been a significant worsening of mental health observed in the young people (N1, L2, L3). In addition, it was felt that the substance misuse patterns had changed, with many more young people getting questioned over cannabis possession, which could be in part because they were smoking cannabis away from their home, especially if they lived with their families, who were now staying at home much more. The patterns of offending and reoffending were also different in other ways under the COVID-19 restrictions, with the retail outlets and places serving alcohol closed (P1, L2, P1).

It was universally acknowledged that it was easier to reach the young people on the telephone than to ensure they attended a meeting in person. However, the views as to the relative value of the in-person versus phone engagement varied (N3, P2).

It was also acknowledged that the switch to less face-to-face contact with the navigators brought about some other benefits (N2).

The phone mode of delivering LINX workshops was in limited use outside of the periods of the COVID-19 restrictions, for example when there were insufficient numbers of clients to run a gender-specific group within the time frame of the clients’ conditional caution. The difference between the phone and in-person delivery of LINX is covered under the theme of Clients below.

Subtheme: overall engagement

It was acknowledged that difficulties engaging the young people was an issue throughout the Gateway project. It was often challenging to get through to the young people on the phone, could take a considerable amount of time to build rapport with the clients, and their continuing engagement was not guaranteed (P1, N2, N3).

Subtheme: engagement with LINX workshops

The engagement with the LINX component of the programme was reported to be a particular issue. The staff commented on the value that many clients derived from attending the workshops, despite apprehension with regards to attending and interacting with other clients (N3, HC1, HT1). The phone call mode of LINX workshops in place during the COVID-19 restrictions or when there were insufficient numbers of clients for a group, undoubtedly led to fewer breaches for not attending LINX workshops, but there was concern that this offered too much overlap with the support already provided by the navigators (HT1).

Subtheme: attitude towards Gateway

This topic of the attitude towards Gateway on behalf of the young people was explored in detail in the interviews with Gateway clients at T2, presented above. This subtheme explored whether the staff felt that the association between Gateway and the police mattered to the young people it was targeting. This question was most relevant to the Gateway police officers, and the impression was that this was indeed possible, but assumed not to be a widespread issue, and little evidence of this tendency came to their attention (HC1, HC2).

Subtheme: time

One of the most often cited reasons for failing to recruit participants was time pressures. This subtheme was naturally closely linked to others, such as that of knowledge, explored below, due to the perception that following Gateway-related processes would be time-consuming, requiring additional time for working out the steps and actions involved. The time pressures on an average shift meant that prioritising was important, and the custody clock (the time period during which a person can be detained, which is 24 hours), for example, was an important priority (R13, R8). Lack of time was seen as a limiting factor in a number of ways: in relation to reading e-mail communications about Gateway, having little time to print off the materials and having a discussion about Gateway with a young person in custody. There was an impression of fearing the bureaucratic side of the recruitment process, especially when not familiar with the processes (R7).

Subtheme: knowledge

Lack of being fully informed about Gateway and associated procedures was frequently cited as a barrier. It was felt that there were many competing demands for attention, and it was challenging to keep up to date (R9).

With the frequently changing guidelines and initiatives, several interviewees cited forgetting about availability of Gateway as an issue. Some mentioned forgetting that Gateway was an option when dealing with an eligible young person, and subsequently receiving an e-mail, alerting them to the missed opportunity (R9).

Some eligible young people may have been dealt with in the usual way, as everyone else, without recalling at the right time, ‘the extra thing that we have to think about’ (R1).

It was common practice for personnel to move internally, with a resultant high turnover, which could lead to temporary unfamiliarity with the requirements of the new post, including the conditional cautions such as Gateway. In addition, an ongoing intense recruitment drive was mentioned, which was linked to the fact that a significant proportion of the workforce was newly recruited, and still learning the ins and outs of the job (R1).

Subtheme: training

All the participants were questioned with regards to receiving training, and a total of five reported having not attended any formal training and some volunteered that they were unaware that such training had been available (R3).

One of the participants who could not initially recall any formal training, checked their electronic records during the interview, which showed that they did, in fact, attend an in-person training session in 2019, which was over 2 years before the interview.

However, the participants reported feeling confident in their ability following the Gateway guidance available on HC Intranet, but were not always aware of the existence of the Gateway project officer(s) (R3).

Subtheme: ‘selling’ Gateway

Not fully understanding what the intervention involved was cited as a barrier to being able to describe Gateway to potential participants in a comprehensive way (R8).

Occasionally, a degree of mistrust, expressed by the young people towards the offer of Gateway interfered with recruiting efforts (R12).

Subtheme: barriers for young people

Some participants felt that the conditions of Gateway may be difficult for some young people to fulfil, for example due to having to take time off work and travel to the workshops. This may have been a factor discouraging some of the recruiters from holding conversations about Gateway, albeit the extent of this is unknown (R9). There were also suggestions that Gateway may be too involved for someone who had much going in their lives and who wanted the ‘easier’, court option (R8).

Subtheme: timely guidance

Participants expressed a preference for timely guidance on whether a particular detained person would have been eligible for Gateway. It was mentioned that the decision-making in relation to eligibility, perhaps, should have even been taken out of their hands. They would have appreciated, for example, an electronic prompt on their computer system, suggesting Gateway at the time of a case being managed (R7, R5). Among all the competing demands on the recruiters’ attention in their daily job, it was very easy to forget about Gateway, in particular, when dealing with eligible cases on an occasional basis (R1).

Subtheme: complexity

Participants reported that the procedure for recruiting into Gateway was easy to find on the HC Intranet, and the process was not overly complicated to follow (R4). Some recruiters, however, commented on operational aspects that they felt were unclear or cumbersome, for example, a confusion of who did what element of the intervention (R10).

Most participants said that the Intranet page for Gateway was easy to navigate, although a small minority had made suggestions on how it could have been improved. Sometimes opinions could be contradictory (R4, R8).

Subtheme: physical locations

The recruitment process was frequently compounded by the features of the premises where the recruitment took place. Where distance divided printers, laptops, custody sergeants, who make the final decision, and custody cells inside a police station, this complicated the process (R8, R2).

Subtheme: randomisation

Randomisation, albeit seen as not unduly complicated to carry out, was in practice a challenging concept, particularly when Gateway was not the eventual outcome. The task was to tell the young people about the programme, which they would only have a 50% chance of getting. If the potential participant was randomised to control, this was disappointing for the young people (R12, R7).

Numbers of eligible young people

During the study planning stage, it was initially estimated by HC that recruitment of the target sample of 334 participants would be possible within 12 months. However, this figure included the numbers of those pleading guilty in court and, in practice, by that stage they were outside of the jurisdiction of the police and hence no longer accessible. This meant that the monthly rates of eligible young people coming through HC in Southampton were much lower than originally anticipated. Several options were considered by the TMG to compensate for this, and the following changes implemented.

Potential participants missed or declining randomisation

This challenge to the study recruitment was multifactorial, with a range of underlying reasons also voiced by a sample of the HC police officers and civilian staff that acted as Gateway recruiters, in qualitative interviews reported in Chapter 3.

Main reasons for non-recruitment of eligible young people:

• Forgetting about, or being unaware of, the Gateway programme and RCT, or misinterpreting the eligibility criteria.

• Time pressures and prioritising other operational tasks during a shift, in preference to following the procedure for Gateway.

• Apprehension about following an unfamiliar process of recruitment and randomisation.

• Being out of date in relation to Gateway training or having never attended training. This could have been because it was not compulsory or not requested by a senior officer or being unaware of training being available (details of the training which was made available for Gateway are provided in Chapter 2, under Recruitment).

• Lack of understanding of Gateway and the difference between randomisation and Stage 1 consent for the study. This may have occurred partly due to lack of training or practical experience, leading to difficulties when explaining Gateway to potential participants or answering questions. This may have contributed to the high rates of study refusal.

• Occasionally, decisions being made that Gateway would not have been in the best interests of the person in question, for example they would have struggled with compliance, or other reasons.

• Other programmes being prioritised, for example Cautioning Adult Relationship Abuse (CARA) for low-risk, first-time domestic abuse offenders.

• Apprehension towards research in the criminal justice system or doubts with regards to its value.

• Lack of awareness of the shopping gift card incentive offered by the UoS for study follow-up or discomfort in relation to discussing such incentive, or disagreement with the idea of an incentive altogether, leading to not mentioning the incentive, which was otherwise anticipated to boost recruitment. As a result, the prospect of taking part may have been seen by potential participants as insufficiently attractive, which may have led to the study being declined.

• The RCT being declined for a variety of other, often highly individual, reasons.

Each non-recruitment of an eligible young person was followed up with an e-mail from the Gateway Project Officer to the colleague who had dealt with that individual, enquiring as to the factors contributing to this. Later, Gateway recruiters’ supervisors were also copied into the e-mail, to increase awareness and response rate. However, these measures were thought to have limited effect, and the rates of missed eligible people continued to be high.

The extent of the barriers to recruitment was more fully revealed in the qualitative interviews with the Gateway recruiters following the closure of the RCT. Low rates of eligible young people, meaning few opportunities to both consider offering the study and discuss it in practice, as well as a high internal staff turnover, were mentioned in the interviews and likely contributed as mechanisms behind many of the above reasons.

Contactability

Low contactability of participants, that is the condition of being contactable, was the most significant barrier to retention in this RCT. This applied both to initial attempts by the researchers to establish contact, before Stage 2, and throughout the follow-up period.

Difficulties contacting participants were due to the following:

• Incorrect, absent or no longer in use contact details.

• No response to initial or subsequent contact attempts.

• Inability to hold a conversation about the study – participants reporting being unable to speak at the time of calls, terminating calls prematurely or asking for calls to be rearranged and subsequently becoming uncontactable.

Missed appointments

There were multiple instances where participants either did not attend in-person appointments (prior to the COVID-19 pandemic) or were uncontactable or no longer available to complete the questionnaire at the time of booked telephone appointments. Approximately 34% of participants at week-4 rising to around 46% at week-16 and year-1 were contacted more than once to schedule an appointment. The follow-up was recorded as a ‘did not attend’ (DNA) if continuing efforts on behalf of the researchers did not result in a completed questionnaire for that time point. Although participants were never questioned as to why they missed an earlier appointment, if contact was subsequently established, some young people offered their reason, which included forgetting, being too busy, falling asleep, having their phone in silent mode, and personal or health issues.

Withdrawals

Most withdrawals occurred while attempting to establish contact following Stage 1. Frequently, a participant would deny any memory of the study (which was not uncommon in general), but also was against continuing the conversation and objected to further contact from the researchers. Withdrawals via text following receipt of a study contact text or letter or after Stage 2 were exceptionally rare.