Scottish Government

CHAPTER TEN Conclusions and recommendations

10.1 This report evaluates the first two years of the ISMS programme in the seven phase 1 LAs of Dundee, East Dunbartonshire, Edinburgh, Glasgow, Highland, Moray and West Dunbartonshire. The report has summarised the policy context for ISMS and the roll-out of the programme across the seven local authorities. The report has attempted to explain the roles of the other key external providers, e.g. Includem, NCH and Serco. Finally, the report has considered evidence from the primary research - principally the surveys and the case studies.

10.2 Below we discuss the main conclusions from our research, attempting to triangulate the evidence (where possible and appropriate) with the findings from each of the research instruments.

10.3 We have based the thrust of this chapter on answering a number of key questions that the Scottish Government put to us in the wake of our research.

Does ISMS work?

10.4 In answering this overall question, we considered a number of related questions, which are detailed below. This is not an easy question to answer as we have examined seven different models of delivery, all at different stages. This is exacerbated by the fact that some of the phase 1 local authorities have very few ISMS cases and that the programme has only run for a short period of time. To prove robustly the extent to which ISMS has produced better outcomes, we would need to conduct a randomised control trial ( RCT), which is seen as the gold standard in relation to testing the effectiveness of interventions. Within an RCT, there would have been a closely matched control group which would have allowed us to compare the outcomes at the end of the evaluation for both groups. However, for practical and ethical reasons, it was not possible to do this in this context as young people who met the same conditions would have to be deliberately excluded from a programme from which they could benefit.

10.5 However, this evaluation has gathered a lot of information on the implementation of ISMS in each area, but outcomes data has not been routinely collected by local authorities and is, at best, patchy. The study has collected case study evidence on outcomes for a relatively small number of young people on ISMS and we have based our findings on this data and additional data that we have collected from SCRA and some of the local authorities. However, outcomes data is not available for all young people who have engaged with ISMS.

10.5 There is little to compare with ISMS in the literature because it is a unique intervention. However, there is some research evidence that suggests that elements of the package do work in terms of tackling offending behaviour, especially the intensive support element. The evidence on the effectiveness of electronic monitoring is more mixed, but leans towards limited or no impact on key criteria, such as reduced recidivism.

10.6 However, as demonstrated below, evidence from the case studies and local evaluation and monitoring work indicate that the ISMS and intensive support service programmes have been effective for a significant number of young people in terms of reduced offending, improved attendance rates on programmes, reduced absconding and reduced substance misuse. However, the quality of the local monitoring and evaluation work is variable. Some areas systematically record details on offending, attendance rates, etc., whereas others do not and their evidence tends to be more anecdotal and less robust.

10.7 Providing young people with intensive support while electronically monitoring them in this way is a unique intervention, and therefore was not set-up with a clear evidence basis. There is some evidence that suggests that elements of the package do work in terms of tackling offending behaviour, especially the intensive support element. The previous literature on the effectiveness of electronic monitoring alone is more mixed, but leans towards limited or no impact on key criteria, such as reduced recidivism.

10.8 Further work would be beneficial in establishing an evidence base on 'what works' in delivering ISMS and intensive support services. In order to conduct robust outcome evaluation of interventions, a high degree of 'programme integrity' at the delivery level is required. 'Programme integrity' requires programmes to have clear and consistent aims and objectives which are linked to the methods being used, and to be delivered by appropriately trained, experienced and soundly managed staff. According to Hollin (1995) it "simply means that the programme is conducted in practice as intended in theory and design".

10.9 At this stage, we cannot assume that ISMS and intensive support service implementation has been conducted in accordance with all of the principles of programme integrity, given the level of variation between and possibly within local areas in many aspects of delivery, and so this aspect of ISMS and intensive support service implementation ought to be addressed if outcome evaluation is to be conducted at any future stage.

Does ISMS reduce re-offending?

10.10 Overall, it is important to bear in mind when considering the impact of this kind of intervention that meta-analysis of research conducted on interventions published between 1964 and 2002. This analysis found that in general programmes with a similar population to ISMS reported a 9% reduction in recidivism for those who had participated in interventions when compared to a control group (Latimer et al 2003). The research which formed part of this meta-analysis was of 'gold standard' and therefore the results of this analysis can be seen as robust.

10.11 Within our research two of the local authorities, Dundee and Glasgow, are systematically collecting data, specifically on ISMS, in relation to the effects on offending behaviour and re-offending. However, we are not able to compare these results to a control group (for the reasons outlined earlier in the paper) and therefore the results detailed below can be seen as showing the reduction only for those receiving the intervention rather than showing a comparison with those who had not received ISMS within the same period.

10.12 In Dundee, the ISMS Team has self reported significant falls in offending behaviour before and after ISMS. The reduction in offending of ISMS cases six months pre-assessment and six months post-assessment was 70%. 12 months pre-assessment and 12 months post assessment (for cases where there is information available), the fall was 52%. These large falls were experienced for a number of categories of offences. This data was collected and analysed by the ISMS Team using police data on ISMS cases.

10.13 In Glasgow, the reduction in the average monthly offending rate prior to and during ISMS was 28%. There was also a 14% fall in the seriousness of offending based on the Gravity Scale developed by the Youth Justice Board and Oxford University (see Vaswani 2007).

10.14 Case file information, reviewed as part of the case studies, indicated, generally, that young people on ISMS reduced their incidence and seriousness of offending behaviours. In addition, there appeared to be a similar reduction in levels of offending recorded in the files of young people placed on intensive support services.

Does ISMS reduce indicators of risk of future offending?

10.15 In Dundee, the reduction in offending behaviour traits is also demonstrated by the reduction in ASSET scores. For those on ISMS, the average ASSET score was 29 at assessment, 20 on coming off the MRC and 20 on ISMS exit (after the throughcare phase). This fall of 31% demonstrates that the risk factors that lead to young people re-offending are reduced by ISMS for these young people in Dundee.

10.16 In Glasgow, there was, again, a more modest reduction, of 8%, in YLS risk assessment scores (from 28.5 to 26.3).

10.17 Other data collected from Includem showed little in the way of improvement in ASSET scores, although behaviour did seem to have stabilised.

10.18 Education attendance also improved for ISMS cases in Dundee, with a 66% attendance rate after ISMS compared to a 37% rate before it. Improved educational opportunities are linked by research with reducing the risk of re-offending.

Does ISMS improve outcomes for young people?

10.19 All of the phase 1 LAs stated their support for the ISMS programme and believed that it was effective in improving outcomes for young people, although it does have its limitations, e.g. it is difficult to get favourable outcomes in cases where problems are more entrenched and many of the young people coming off the MRC are 16 years-old and resources are not as available to help them afterwards.

10.20 Local teams believed that ISMS had provided significant improvements in the self-esteem of young people during and after the programme. They are better motivated and have increased likelihood of employment, through better links to employment opportunities.

10.21 In our stakeholders' survey ⁵⁵, in comparison to the other measures used within the Children's Hearings System, ISMS was generally viewed to be as at least as effective as these measures. ISMS was viewed as more effective than home supervision by 71% of the participants. ISMS also compared favourably with secure accommodation, accommodation other than secure, and intensive support (without MRC). Compared to all three of these other measures, at least half of the respondents said ISMS were more/much more effective.

10.22 A similar story emerged when we asked stakeholders to compare ISMS with measures available in the adult system, again with at least half of the participants viewing ISMS as more or much more effective than these other measures.

10.23 In our survey of Panel Members, the majority of respondents judged ISMS as being no less effective than other measures open to them. ISMS compared least favourably with secure accommodation, however, almost two-thirds of respondents still considered ISMS to be at least as effective as secure accommodation. More than half of respondents said that ISMS was more effective than intensive support alone.

10.24 In our case study research, some parents and foster carers responded that they felt that the increased amount of time within the home environment created the opportunity to re-establish their relationships with each other. The MRC and associated support was seen by some as having greatly assisted parents in being able to re-assert their rules.

10.25 Breaches of the conditions of ISMS were also commonly reported in case files. However, breach is not necessarily failure if it helps young people to engage with interventions, when they had not previously engaged, and if it leads to a drop in their risk factors. For those young women who received ISMS because of absconding, the more substantial breaches were associated with them staying out in the community, often when they were abusing substances and, potentially, placing themselves at risk. It is not clear in other cases if the breaches have any relation to other outcomes.

Who is ISMS being targeted at?

10.26 The data from the phase 1 areas suggests that ISMS is predominantly targeted at males over 15 years-old. It also tends to be targeted at those young people who are a risk to others, i.e. Offenders, rather than those who are a risk to themselves or a risk from others, although ISMS is still used for a number of these cases. It was felt by the ISMS programme co-ordinators in a number of areas that an MRC may cause some of these vulnerable young people unnecessary and additional distress. If the MRC is seen as the punitive part of ISMS (and it is seen in this way by some professional staff, young people and their carers), then punishing young people for being at risk when they are committing no offence seems very harsh to many professionals. However, our case study research covered a number of cases where an MRC had been given because young women had been absconding.

10.27 Although ISMS appears to be targeted at young male offenders over 15 years-old, information drawn from the case studies presented a relatively consistent pattern of complex needs in ISMS cases. Both the young men and women considered, generally, have histories characterised by multiple referrals to the Reporter for care and protection and offending grounds. Young men and women placed on ISMS, generally, were found to have been referred earlier in their lives and to have had a history of contact with the Children's Hearings System resulting in Supervision Requirements having been made several years prior to the ISMS. Issues such as substance misuse, truancy and absconding from home appear in the backgrounds of the young men and women.

10.28 The case studies did demonstrate differences in the use of ISMS for young women and young men. Concerns around young women, placed on ISMS, circle around placing themselves at risk through absconding, substance misuse and fears about personal safety (which usually translate into fears about sexual exploitation while out in the community). These young women also have referrals to the Reporter for offending, as well as other offences listed in their case files, but the files maintain an emphasis on reducing their risk from others and not offending issues as the major concerns. For young men placed on ISMS, offending behaviours appeared predominantly to have influenced the use of the disposal.

Who do ISMS work for and how?

10.29 Those leading the programme in Dundee believe that ISMS has demonstrated its ability to tackle the behaviour of those that are a risk to others, e.g. Young male offenders, but it has not yet demonstrated this ability to extend beyond this to other cases. However, the ISMS Team argues that this position may be changing as they and other partners become more comfortable and confident about assessing and managing risk.

10.30 In our survey of Panel Members, ISMS was generally considered as being more appropriate than secure accommodation for persistent offenders and for younger children. Secure accommodation was generally preferred for serious offenders and for cases where there was a risk of self harm or absconding.

10.31 The case studies indicated that most of the positive outcomes attributed to ISMS were perceived by social workers and carers to be a consequence of the range of interventions and access to resources that were the intensive support facet of the package, rather than the MRC. Key elements were perceived to be the 'forced' engagement of young people with services and support, and the length and consistency of support. However, the MRC was believed in having a positive contribution in certain types of cases, e.g. in allowing young people to develop a set of credible reasons for not engaging with offending peers. Some of the young people interviewed appeared to reduce risky behaviours because of the potential consequences of not meeting the ISMS conditions.

10.32 Based on the interviews reviewed in the case studies, the ISMS disposal facilitates access to protective factors in the community, such as engaging with college placements and building relationships with family members and friends. In terms of community accessing resources and maximising social inclusion, the intensive support packages appear to have increased young people's use of education, employment agencies and health agencies. ISMS, therefore, appears to work by allowing young people access to resources that could act as protective factors (which, in turn, reduce risk), and to receive structured interventions that also serve to reduce risk.

What does a successful ISMS programme look like?

10.33 Under Scottish Executive guidance, an ISMS disposal involves the young person receiving intensive support with an MRC for a period of time (which can vary). This is preceded by an assessment phase (of around six weeks, where intensive support is provided) and post- MRC intensive support of the equivalent period for which the MRC was applied. In each of the LA areas, the average time on the ISMS programme and the MRC varies, at 17-38 weeks for the programme and 10-17 weeks for an MRC. These averages seem to be in line with the initial guidance material. The curfew hours also vary, but are normally from evening to early morning.

10.34 The main external provider of ISMS services is Includem in five of the seven phase 1 LAs (Glasgow, Dundee, East Dunbartonshire, West Dunbartonshire and Edinburgh). Highland Council uses NCH, while Moray does not employ an external agency, but delivers these services through an in-house team. This support is complemented by other services such as education, social work, addiction services (where appropriate) and mental health services (where appropriate).

10.35 It is, arguably, Glasgow that has the most developed ISMS programme (it is the most established and has had the biggest throughput) and has the most robust evidence base to support its positive impact. There are seven key elements in the Glasgow ISMS model.

1. Social Work Youth Justice Teams, which provide case management (these workers are trained and in place).
2. The ISMS Project Team are responsible for co-ordinating assessments, accessing services and providing some limited direct work.
3. Includem, providing direct care, out of hours and crisis/respite care.
4. Educational/vocational training provision that will emphasise individual approaches to learning (their recruitment process has started). A specialist education resource is established at the ISMS base.
5. Forensic Child and Adolescent Mental Health Services ( FCAMHS) - this service is being enhanced to offer specialist services or will procure more appropriate services from other city services.
6. Youth addiction services - support and intervention to tackle substance misuse.
7. Restorative justice interventions - where these are considered appropriate.

10.36 ISMS delivery in Glasgow has focused on case management to fit complex packages around the needs of individual children in a tailor-made way. This happens in the other areas as well.

10.37 The average weekly interventions by agency at the start of an ISMS programme in Glasgow in 2006-07 were:

10.38 The average total weekly intervention is 25.7 hours. The average MRC is for 70 hours per week.

10.39 However, other areas tend to have higher volumes of Includem or NCH support, normally around 20-25 hours per week on average, and of education support. Total weekly interventions are at around 40-50 hours per week. The elements of the programme are proportionally the same in other areas as in Glasgow. Glasgow intended to have a 50 hour per week programme, but there has been difficulty in getting education provision for the young people up to the levels initially expected. The LA also felt that the initial 50 hour target was too ambitious. Dundee is also a little different, with much wider variations in the timings for particular interventions, which are dependent on assessment of young people's needs.

10.40 There is no clear evidence on what components an ISMS programme needs to have to be 'successful' or what the level of these interventions needs to be in terms of intensity.

10.41 Support services provided by Includem (in five of the LA areas) vary depending on the needs of the young person and their circumstances. At the heart of the Includem model is one-to-one support based on a professional relationship ('pro-social role modelling'). Includem's approach is about supporting and challenging a young person's behaviour at his or her home through these established relationships. This usually involves providing support in the community, addressing offending behaviour and other problems the young person may have, and providing 24-hour crisis support. It is a mixture of informal befriending, e.g. trips to leisure centres and fast food outlets, in order to gain engagement with the young person, before moving on to more structured work around, for example, offending behaviour, substance misuse or relationships. This often involves work with families/carers and key support workers as well as the young person themselves. Service provision by NCH in Highland and the in-house team in Moray follows a similar model.

What are the factors contributing to programme success?

10.42 There appear to be a number of major factors that need to be in place in order to implement a successful ISMS programme:

1. The programme must be driven forward from an early stage by a suitably skilled individual who takes responsibility for the organisation and implementation of the key elements that need to be in place to provide a service to meet local need. S/he should be appropriately supported to do this.
2. There needs to be high level management support for the programme within the LA.
3. Effective partnership arrangements need to be built early on, especially involving Education, Social Work, external providers and others as appropriate. The different agencies and workers need to be aware of their and other's roles and responsibilities.
4. There needs to be a supply of suitable accommodation for young people in an area.
5. The programme needs to be marketed effectively, especially to the police, the Courts and the Children's Hearings System so that ISMS will actually be used. Its profile also needs raised in the wider community so that people are aware of it and what it is intended to do.
6. There needs to be an effective link-up with secure providers to ensure that those young people in secure are assessed for ISMS and that work (such as assessment work) is not being duplicated.
7. The multi-agency case conference approach to screening and assessment seems to be the best way in which to gather all of the opinions of the professional staff involved in a case and discuss and debate the issues surrounding a young person and what they need.
8. There needs to be flexibility in programme construction and delivery so that the needs of different types of young people are met.

10.43 Difficulties in achieving all of these requirements has led to delays in implementation of ISMS and a lack of throughput of ISMS cases in certain areas during phase 1.

10.44 In our survey of stakeholders, we asked them what they considered to be the most effective aspects of the ISMS programme. The crisis response mechanism had the highest rating (considered by 84% of respondents to be either effective or very effective). This was closely followed by the providers Includem/ NCH. Our survey of Panel Members produced similar results.

10.45 The services delivered by Includem/ NCH were singled out by the large majority of interviewees in the case studies as being the most positive and effective aspect of the intensive service. Such commendation was also found in many case files and noted in some Children's Hearing reports. There appear to be a number of similarities between case study findings and the wider evidence base on good practice in the supervision of people with complex needs and risk in the community. The Pro-social Modelling approach contains at its core the key elements of engaging with the client's motivation while simultaneously building on the relationship between the worker and the client. The quality of the relationships that developed appeared to involve themes of trust, empathy, a non-blaming approach, holding optimistic views, and being able to articulate the client's and family members' feelings and problems that are entirely congruent with the Pro-social Modelling approach.

10.46 The Programme Manager in Dundee is of the view that ISMS has been a significant success for a number of reasons. Generally, these were shared across the LAs, however, this was the most complete set of reasons given by a Programme Manager in any of the areas and Programme Managers in some of the other areas may not be in full agreement about all of these conclusions.

• The MRC - young people are saying that the MRC enables them to avoid their peer group and parents believe that it acts as a temporary means to control the whereabouts of their children that allows the ISMS Team to work with the young people and their parents to provide them with the skills with which to impose this control when the MRC ends.
• Education - as a means of bringing structure and direction to a young person's life.
• ISMS being an alternative to secure - so that the young people have the responsibility and commitment to make the programme work.
• The crisis hotline run by Includem - parents say that during periods of hostility, the helpline is very useful to express concerns about behaviours and discuss methods to resolve the hostility with the potential of Includem visiting to support.
• The flexibility of Includem support - being able to increase or decrease contact hours depending on changing circumstances.
• Feedback from Serco on non-compliance - which assists the ISMS Team in terms of risk assessment and management (although there were issues about the responsiveness of Serco in a number of other areas).
• The temporary crisis accommodation facility - which can ease tensions at a young person's home and allow the ISMS Team to continue to work with the young person and their parents to help to get the programme back on track.

What are the barriers to programme success?

10.47 There have been a low number of ISMS cases in phase 1, much lower than initial estimates. In some areas, the estimates were not achieved because insufficient priority was given to the ISMS programme and/or because of ideological concerns about the nature of electronic monitoring. It does seem that the number of cases was over-estimated at the outset, but it also appears that the high use of intensive support services has impacted on the use of ISMS.

10.48 There is an amount of confusion over the guidance on who ISMS should be targeted and this has also affected throughput, with disputes between the different agencies as to whether ISMS is suitable for particular cases. This has been particularly evident in Moray and, given that the agencies need to agree on ISMS before it can be implemented, this sometimes leads to ISMS not being sought. The Scottish Government states that ISMS is intended for those who meet Section 70(10) conditions and, therefore, act as a disposal for a relatively small number of young people. It appears doubtful whether this guidance has been stuck to in all areas throughout the evaluation period, especially with the relatively high number of intensive support service cases in relation to ISMS cases. However, there does seem to be a case for using ISMS (or intensive support services) as a preventative measure for young people who may meet these criteria in the future.

10.49 In some of our case studies, it appears that interventions provided by Includem and other agencies, such as substance misuse services, did not adequately address areas of need that are specifically related to the risk of further offending. It appears, in these cases, that the quality of interventions provided would not meet recent good practice guidelines detailed in Getting it Right for Children and Young People Who Pose a Risk of Serious Harm (Scottish Government, 2008). ⁵⁶

10.50 Lack of training for stakeholders is also a barrier in terms of lack of awareness and understanding among key professional groups. In our survey, just over half (53%) of the stakeholders received training, meaning that a large proportion of stakeholders had received no ISMS training. Of those responding to the survey, none of those working in secure accommodation or children's units; legal professions; and ASB services had received any ISMS training. It is likely that training has been targeted at those most core to service delivery, however, it still appears that some keys groups of professional staff have still to receive training, notably those working in secure accommodation and children's units.

10.51 Education is a key element of ISMS service provision as many of the young people on the programme will have been out of mainstream education for some time. Achieving levels of educational attainment and/or qualifications are key to future employability as well as in improving self esteem. Direct employment of teaching staff would seem to be the most effective way of providing this, otherwise a key need is not addressed and it is more difficult to keep a young person occupied on a programme. Some areas have undertaken this. Most areas now have the level of education provision that they require, but it has been a challenge to provide it. Other areas, e.g. Moray, are still struggling to provide education at anticipated levels.

10.52 Access to suitable housing and accommodation appears to be a major problem across the phase 1 LAs and means that young people who meet the Section 70(10) criteria do not receive ISMS, as there is nowhere suitable to tag them to, or because it is difficult to maintain them on the programme. The case study research highlighted that being subject to ISMS conditions also appeared directly to have a negative effect on some people, as it was reported that some residential options were taken away because the young person was being or had been electronically monitored.

10.53 Relationships between the core services delivering ISMS (including education, social work and intensive support providers) appear strong, but more work has to be done in integrating other services, notably Housing, Mental Health Services, Employment Services and Addictions Services. In our survey of stakeholders, accommodation services were viewed by 28% of respondents to be either ineffective or very ineffective and a large proportion of respondents (38%) viewed it as neither effective nor ineffective. Employment services, mental health services and addiction services were the other parts of the programme that scored were considered effective by less than 50% of respondents.

How are intensive support services being used?

10.54 The focus of our evaluation is ISMS, not intensive support services, although we have asked LAs for high level figures on intensive support services and included intensive support service cases in our case studies to act as a broad comparator group. However, the evidence gathered on intensive support services alone is limited.

10.55 Intensive support services packages essentially provide the intensive support elements of the ISMS package but without the MRC. Having said this, the nature of the programme may also be different, e.g. less intensive, because intensive support only cases in general, although not always, tend to have lower levels of risk and need than ISMS cases. This fits with the concept of tailoring packages of care and need to the individual young person.

10.56 There have been a large number of intensive support only cases, three times as many as ISMS cases. Many of these cases were considered unsuitable for ISMS because the intensive support that they received during the assessment phase rendered the MRC as no longer necessary. There were other problems in moving many cases on to an MRC, such as a lack of suitable accommodation or difficulties in the young person's family relationships.

Who is receiving intensive support services?

10.57 As highlighted above, it is difficult to answer this question definitively because we have not asked for a detailed breakdown of intensive support service cases (it is not part of the remit of the evaluation). In theory, intensive support service cases should have lower levels of risk and need (because they do not have to meet the Section 70(10) criteria). It also appears that young people at a risk to themselves or from others are more likely to receive intensive support services (rather than those who are referred to the Reporter on offending grounds). However, the case studies demonstrate that there are cases where intensive support service is received by high risk offenders (because there is nowhere suitable to tag them to) and for those at a risk from absconding (because practitioners feel intensive support service is more appropriate).

What are the phase 1 areas spending their ISMS money on?

10.58 The LA spending patterns are all very different. This is not surprising as each LA was delivering its own model and had different requirements for providing ISMS from the start because of different existing provision and resources.

10.59 Despite the very different proposed models for delivery, different infrastructure requirements, etc., the expected unit costs per ISMS case and per assessment in the areas was broadly similar (within a range of £38,000 to £80,000 for ISMS cases).

10.60 Total expected costs were also very similar to total actual costs in each area, but, due to the much lower than anticipated use of ISMS, actual unit costs are considerably higher than expected unit costs for all areas, although the difference is noticeably smaller for West Dunbartonshire.

10.61 ISMS has cost £10,944,404 to roll-out in the seven phase 1 LAs in a two-year period. This equates to around £5.5 million per year. To provide a proxy of how many of the likely ISMS cases that these areas account for, we used children referred on offence grounds as persistent young offenders for 2005-06 (from the SCRA Annual Report 2005-06). These five areas make-up 39% of these cases. Based on this, we would estimate the cost of national roll-out of the ISMS programme to be around £14 million per year ⁵⁷.

Recommendations for future research

10.62 Below we list a number of recommendations for future research, particularly targeted at assessing more clearly the impact of ISMS and of providing accurate unit costs of the programme.

1. The Scottish Government should co-ordinate future research into the effectiveness of ISMS/intensive support services using processes that are standardised across LAs that allow the statistical analysis of this relatively small population. The research should be designed to capture sufficient numbers in each LA to make comparisons possible. A quasi-experimental method would permit investigation into the effect of the MRC itself, when combined with multivariate statistical analyses that would control for confounding factors that may influence the outcomes.
2. The Scottish Government should work with the LAs to devise local monitoring and evaluation work consistent across the LAs that will collect data on agreed outcome measures of those on ISMS/intensive support services.
3. Future research could include a closely matched control group to allow the effect of the interventions to be assessed in comparison to another group of young people who did not receive these interventions.
4. Future research could usefully include a longitudinal design, where repeated and consistent data from the same individuals are collected over a longer time period. This methodology allows findings to be disentangled through their temporal relationships and developmental sequences of events to be identified.
5. The measuring of the unit cost of the ISMS and intensive support service programmes should be the subject of further research, which attempts to separate out data and costing on intensive support service and ISMS as discrete exercises.