Research and analysis

Violence Reduction Units 2022 to 2023

Published 20 December 2023

Applies to England and Wales

Executive summary

In 2019, the Home Office announced that 18 police force areas (PFAs) would receive funding to establish (or build upon existing) Violence Reduction Units (VRUs) as part of the Serious Violence Fund. In 2022, a further 2 PFAs received funding to establish VRUs. The aim of VRUs is to lead and coordinate a preventative, whole-system approach to violence reduction, which comprises:

• multi-agency working
• data sharing and analysis
• engaging young people and communities
• commissioning (and delivering) evidence-based interventions

Informed by the findings and recommendations from previous VRU evaluation reports, there was a particular policy focus for 2022 and beyond to:

• further strengthen multi-agency working
• encourage and support the sustainability of VRUs
• improve the quality and granularity of data accessed, and the analysis of this
• commission and develop the evidence for high-impact interventions

The Serious Violence Fund also covers Grip (formerly Surge) activity, which is enforcement focused. Combined, VRU and Grip funding represents a total investment of £315 million over 4 years to tackle serious violence (SV). PFAs with the highest (historical) counts of violence – specifically, hospital admissions resulting from violent injury with a sharp object – were selected for funding. The Home Office commissioned: Ecorys UK; Ipsos UK; the University of Hull; and the University of Exeter to evaluate VRUs in their fourth year of operation (April 2022 to March 2023). The aims of the evaluation were to:

• estimate the impact of VRUs (and SV funding) on violence
• assess the progress made, particularly in the last year, by VRUs towards a whole-system approach

The mixed-methods evaluation included desk-based research, qualitative research with a range of stakeholders in each VRU, and quasi-experimental designs.

Key findings

There were no statistically significant impacts (at the 95% confidence level) on the primary SV outcomes of hospital admissions for sharp object violent injury or homicides. However, the direction of impact estimates for these outcomes did indicate a reduction, and we note that statistically significant impacts are more difficult to detect with these small count outcomes. Regarding secondary outcomes, there was a statistically significant reduction in hospital admissions for any violent injury^{[footnote 1]}. Since funding began, an estimated 3,220 total (or 8 per 100,000 persons) hospital admissions for any violent injury had been prevented in funded areas. Analysis focusing on just violence ‘hot spots’ (top 5% within VRU areas) also indicated a potential impact on police recorded violence with injury and violence without injury offences.

The statistically significant reduction for violence without injury offences at a whole-programme-level that was observed in previous years evaluations was not sustained. However, the wide (and overlapping with previous years) confidence intervals for these estimates indicate a high degree of variability in the data. As such, we cannot say that the results for this outcome reflect a real change in violence (see Section 1.4.1 and Section 1.5).

Overall, the findings this year are encouraging, showing that a statistically significant reduction in more serious forms of violence has been achieved. This represents progress compared to previous years and is consistent with previous research showing that impact on the most serious violence may take longer to materialise.

Building on the good progress made in previous years, VRUs continued to demonstrate progress and showed signs of maturing and becoming embedded in local responses to prevent violence.

Multi-agency working:

• overall, VRUs have continued to make good progress towards effective multi-agency working; this includes evidence of VRUs maturing and becoming more effective over time. Variation in progress can be attributed to the range of partners meaningfully engaging and sharing responsibility
• the introduction of the Serious Violence Duty^{[footnote 2]} and 3-year funding were generally viewed as an opportunity to strengthen partner buy-in, commitment and focus. The match-funding requirement was met by all VRUs, with many achieving substantially more than the 10% required for 2022 to 2023
• effective approaches to multi-agency working centred around building and embedding strong working relationships with partners, having structures in place that enable effective and multi-agency working, and having a clear vision and buy-in from key operational and strategic partners
• challenges to effective multi-agency working included varied engagement from some partners, which included less relative engagement from health and education partners in some VRUs, and uncertainty around future funding. These challenges are consistent with those evidenced in previous years’ evaluations

Data sharing and analysis:

• building on the good progress made in the first 3 years of the programme, there was some evidence of continued progress towards effective data sharing and analysis in 2022 to 2023. This included increased usage of data insights to commission and target interventions
• progress was most evident in VRUs that had continued to invest in systems, such as data dashboards, and could access more and better quality data to improve the VRUs’ and partners’ understanding of those at risk
• however, there was a wide range of variability across VRUs. This included many still facing challenges including access to granular data (in particular, health data), and ongoing efforts to increase data sharing and standardise data capture
• effective approaches to (or enablers for) effective data sharing and analysis included having appropriate points of contact in partner agencies to facilitate data sharing, the use of data governance and technological solutions (for example, data dashboards), and ensuring sufficient analytical capacity/capability within the VRU
• key common challenges included (continued) difficulty accessing health, probation, and individual-level school data, inconsistencies in data quality and granularity, and limited analytical capacity and/or resources

Community and young person engagement:

• progress towards effective engagement of young people and communities varied greatly across the VRUs. More than half of VRUs demonstrated good progress, and, importantly, improved progress based on previous years where things had moved more slowly, relative to other elements of the whole-systems approach
• progress was most evident in VRUs where there was also strong multi-agency working, which could serve as a facilitator for more effective engagement
• effective approaches to (or enablers for) effective engagement included harnessing the 3-year funding announcement to encourage meaningful and longer-term engagement, and co-designing interventions
• creating a consistent feedback loop between community engagement and VRU activity was an ongoing aim and challenge

Commissioning and delivering interventions:

• VRUs continued to make progress towards strategic and evidence-based commissioning and delivery of interventions. This included evidence of VRU-commissioned interventions delivering on their intended early outcomes for young people
• there was variation both in the extent to which VRUs engaged partners in commissioning decisions, and in the specificity with which stakeholders considered interventions as ‘high impact’
• the commissioning and delivery of interventions is most effective when VRUs combine 4 key perspectives: data insights, partners and intervention-level or VRU-wide monitoring (to understand the issues, the need and the gaps); the independence of community-based or third-sector projects; the specialist resourcing and scale of statutory providers; and the strategic perspective of the VRU as a network
• although VRUs demonstrate more effective use of data and evidence to identify areas, target groups and commission interventions, there is a challenge then to operationalise this into engaging those most in need

Recommendations

Where appropriate and feasible, key recommendations for VRUs to consider include:

• develop a clear, shared vision that is, as far as possible, driven by the best available evidence, informed by local priorities and broader government strategies, and ensures buy-in from key partners
• develop and maintain strong relationships: Focus on relationship building (using a people-centred approach), open communication and engaging a wide range of partners
• secure long-term funding and strategic planning: Encourage longer-term commitments from partners through stable funding and strategic planning that enables the development of multi-year interventions. Leverage the SV Duty to increase partner engagement and ownership
• establish and maintain effective structures: Develop structures, such as local partnership structures and embedded local VRU coordinator roles, that facilitate cooperation and coordination. Consider the hybrid VRU model for its potential benefits in efficiently leveraging existing local partnerships and maintaining central decision-making
• invest in data systems and dashboards: Utilising data dashboards can help VRUs to visualise, analyse, and monitor trends and patterns in violence. This can enable data-driven decision-making and promote better communication among partners. Investing in such systems can lead to more efficient and effective data management
• linked to the above, increase analytical capacity: Invest in personnel and resources dedicated to data analysis, including training and support for existing staff. This will ensure that VRUs can effectively interpret and utilise the data they collect
• improve access to health and education data: Work closely with health and education partners to overcome barriers to data sharing, whilst maintaining necessary privacy and data protection standards. Access to this data can provide a more comprehensive understanding of violence and inform targeted interventions. Harness partner relationships and the SV Duty to overcome barriers to accessing this data
• continue to enhance evaluation processes: Use both quantitative and qualitative evaluation methods to provide a richer understanding of the impacts of interventions. Sharing evaluation findings can help identify areas of need and facilitate learning among partners
• continue to engage communities and young people, and work with them and partners to ensure this is with the right groups and considered meaningful for all involved. Ensure that the core VRU team includes those whose experience or perspective, not just their role, enables them to identify how best to achieve this. Ensure a mobilised data and evidence base for more effective engagement with those who are most in need
• work with partners to ensure high-impact interventions are understood as being high impact and of local importance. This could include sharing evaluation/monitoring data to demonstrate the impacts and/or assuring them why they will lead to the intended impacts. This could be supported with the Youth Endowment Fund (YEF) toolkit, and may support wider buy-in and funding commitments to support such interventions in the longer term
• linked to the above, demonstrate to partners (and others) that commissioning decisions consider data insights, and local needs, knowledge and resources

Alongside supporting VRUs with the above recommendations, specific recommendations for the Home Office include:

• continue to support the sustainability of VRUs beyond the current funding cycle. This could include, at a minimum, funding the core function of VRUs (that is, excluding interventions) to enable a long-term approach to violence reduction
• linked to the above, consult with VRUs on their current structure and operation. This includes consideration for the sustainability of the current model
• continue to support the analytical capabilities of VRUs through (ongoing) work on data sharing and analytics platforms, and guidance on the appropriate and effective use (and limits) of insights gained from data
• at a national level, continue to engage with and encourage collaboration across different government departments to support the VRU programme, for example:
  • establish stronger links with the Office for Health Improvement and Disparities, NHS England, NHS Wales and Integrated Care Boards to improve access to health data and improve the wider engagement of health partners at a local level
  • work collaboratively with the Ministry of Justice (MoJ), Youth Justice Board, and Department for Education (DfE) to share learnings, understand overlap and optimise VRU delivery alongside other programmes (for example, MoJ Turnaround and DfE SAFE)
• support and emphasise the importance of high-quality and accurate monitoring data. VRUs being able to demonstrate the impact of interventions will be increasingly important to conversations with local partners as match-funding requirements increase
• continue to work with YEF and VRUs to support the testing of promising interventions and share the learning from this. This may include the development of an evaluation tracker, where all VRU-funded evaluations are stored

1. Introduction

This report presents findings from the evaluation of Violence Reduction Units (VRUs) in the year ending 31 March 2023, the fourth consecutive year of Home Office funding. The findings are primarily based on evidence collected in the 2022 to 2023 funding year but build on context and understandings developed in previous years’ evaluations. Data sources include quantitative data and interviews with professional stakeholders (in strategic, operational and frontline roles), community representatives and young people.

The Home Office has funded VRUs in 20 Police Force Areas (PFAs):

• 18 VRUs since the financial year 2019 to 2020
• 2 additional VRUs this financial year 2022 to 2023^{[footnote 3]}

The government established VRUs in 2019 to primarily provide leadership and strategic coordination of all relevant agencies to support a ‘whole-system’ approach to tackling serious violence (SV). The Home Office commissioned this VRU evaluation in April 2022. The report builds on previous years’ evaluations, which were conducted by the same research team comprising: Ecorys UK; Ipsos UK; Professor Iain Brennan of the University of Hull; and Professor Mark Kelson of the University of Exeter.

This chapter provides the policy background to, and programme-level Theory of Change (ToC), for VRUs, the evaluation aims and an overview of the evaluation methodology.

1.1 Policy background

In the summer of 2019, the Home Office announced that 18 PFAs would receive funding to establish (or build upon existing) VRUs as part of the Serious Violence Fund announced by the Treasury earlier that year. The Home Office selected the areas by the levels of SV (based on hospital admissions data) experienced from 31 March 2016 to 31 March 2018. The VRUs received subsequent funding on a rolling annual basis. Key developments going into the financial year 2022 to 2023 included funding being awarded on a 3-year basis, to enable longer-term planning within VRUs, and the establishment of 2 new VRUs in Cleveland and Humberside PFAs.

The core aim of a VRU is to provide leadership and strategic coordination of all relevant agencies, to support a whole-system approach to prevent SV. This approach is advocated for and informed by work by the World Health Organisation (WHO) and Public Health England (PHE)^{[footnote 4]}.

Key elements of the whole-system approach, which align with the WHO’s public health approach to violence prevention, include:

• multi-agency working to enable an efficient and effective response to violence
• accessing and sharing data to ensure the response targets those in need
• engaging with communities and young people to guide the response
• commissioning and delivering evidence-based interventions to effectively support those in need

Informed by the findings and recommendations from previous VRU evaluation reports, there was a particular policy focus for 2022 and beyond to:

• further strengthen multi-agency working and, linked to this, introduce the SV Duty and VRUs role within this
• encourage and support the sustainability of VRUs. This included a new requirement attached to Home Office funding that local sources should match at least 10% of funding. The requirement will increase in subsequent years
• improve the quality and granularity of data accessed and their analysis. This included improving the alignment between identified needs within VRU areas and the VRUs’ response to these
• commission and develop the evidence for high-impact interventions. This included a new requirement for at least 20% of VRU intervention budgets to be spent on interventions considered as ‘high impact’ by the Youth Endowment Fund (YEF). The requirement will increase in subsequent years

As well as VRUs, the Serious Violence Fund also includes the Grip programme. All 20 PFAs received Grip funding to roll out targeted, analysis-driven policing operations in the highest crime hotspots. The Serious Violence Fund represents a total investment of £315 million over 4 years.

There are also wider connected activities and funding in this space. For example, YEF and work to disrupt county lines operations. VRUs work closely with YEF and police teams to access additional funding and, as far as possible, seek to complement concurrent activity.

1.2 VRU Theory of Change

Figure 1.1 shows the VRU programme-level ToC. The ToC sets out the intended outcomes and impacts of VRUs and the causal pathways (indicated by arrows) to reach these through the inputs, activities and outputs. Activities and their anticipated outputs and outcomes have been structured horizontally by the key elements of the whole-system approach.

Whilst each element comprises distinct activities, outputs and outcomes, each is complementary and collectively contributes to the anticipated (longer term) impacts of VRUs. Multi-agency working is considered a key facilitator of the other elements.

The team updated the ToC based on funding and programme changes, and findings from the 2021 to 2022 national evaluation. Key points to note are:

• updated inputs included new funding amounts, capturing ‘hybrid’ VRU models (see Chapter 3), and recognition of the match-funding requirements, 3-year funding confirmation and the SV Duty
• for the whole-system approach, elements of multi-agency working, data sharing and analysis, and interventions, there were limited changes to the activities, outputs and outcomes
• initially, engagement activities were expected to raise awareness of the VRU among young people and the community, leading to improved perceptions of the VRU and its partners. However, past evaluations did not show evidence of this. Interviewed stakeholders prioritised meaningful engagement to inform and co-develop VRU activities with those most affected by violence. Thus, the ToC was updated to reflect this, following a contribution analysis approach
• the cost-of-living crisis was added as a contextual factor in the ToC in recognition of poverty/inequality being a driver of violence/crime

Figure 1.1: VRU programme-level ToC

1.3 Evaluation aims

The Home Office commissioned the evaluation team for a fourth consecutive year to conduct an impact and process evaluation of VRUs. The broad aims of the 2022 to 2023 evaluation were to:

• estimate the overall impact of SV funding on violence outcomes and, as far as possible, understand the specific contribution of VRUs to reductions in violence
• assess the progress VRUs are making towards the different elements of the whole-systems approach to violence reduction, and consider the impact of VRUs on wider (more qualitative) outcomes related to embedding a whole-systems approach to violence reduction

A programme-level ToC (Section 1.2) and a linked evaluation framework underpinned the evaluation.

1.4 Evaluation approach

The following sections summarise the evaluation methodology. The approach was similar to the 2021 to 2022 evaluation, which is helpful for understanding progress over time, but included additional methods of data collection and recognition of the policy developments and focus detailed in Section 1.1.

1.4.1 Estimating the impact on violence reduction

The evaluation team estimated the impact of SV funding for the following primary outcome measures:

• hospital admissions for sharp object violent injury
• homicides

Reliable data covering knife-enabled SV – the other key outcome specified by the Home Office for VRUs – is only available from 2019 and does not cover all PFAs, so was not suitable for impact evaluation (ONS, 2021).

Impacts were also estimated for secondary outcome measures where, recognising the whole-systems and preventative approach of VRUs (see programme-level ToC), the evaluation team deemed it reasonable to explore potential impacts on the following wider forms of violence:

• hospital admissions for any violent injury
• police recorded violence with injury
• police recorded violence without injury, which are mainly assaults not resulting in injury
• police recorded possession of weapons offences, as an indicator of weapon carrying

Data on all outcomes covered all PFAs and, except for homicides, were accessible up to December 2022. Homicides data was available up to September 2022.

Whilst we advise some caution around interpreting police recorded outcomes where a change could reflect changes in the reporting, detection and recording of crimes, their inclusion aligns to the preventative focus of VRUs and wider SV funding. This was supported by existing evidence of hotspot policing (a key Grip activity) and preventative interventions (commonly funded by VRUs) demonstrating impacts across a broad range of offences (Braga et al., 2019; Gaffney et al., 2021a and 2021b).

Homicide data are taken from the Home Office Homicide Index, which contains detailed record-level information about each homicide recorded by the police in England and Wales and, when published annually by the Office for National Statistics (ONS), are designated as National Statistics. The homicide data analysed in this report use unpublished information extracted from the Homicide Index on 2 December 2022. Whilst data quality is thought to be high, the team has not reconciled this data with forces and there are some homicides where the Home Office is waiting for police forces to submit updated data.

Hospital admissions data are from the NHS Hospital Episode Statistics, which contain volumes of admissions recorded by patient home address and cause code and, when published, are defined as National Statistics. The hospital admissions data analysed in this report use unpublished management information which has not yet undergone the formal badging process, but is produced using similar principles and methodology.

The quasi-experimental design (QED) used to estimate impacts was synthetic control methods (SCMs). SCMs compare trends in outcomes following an intervention in treated areas (SV funded) to a synthetic control group. The synthetic control group is a weighted average of comparator areas (non-SV funded). Comparator areas that are most similar to treated areas on pre-intervention outcome trends receive a higher weighting. This approach, alongside population adjusting outcomes of interest and, where possible, focusing on smaller (than PFA) more comparable areas, allowed for comparisons between SV-funded and non-SV-funded areas that had similar outcomes trajectories (prior to funding), which were more appropriate than simple comparisons (without weighting or adjustment).

Exploratory analysis of violence ‘hotspots’, alongside qualitative insights from VRU stakeholders, were used to understand the specific contribution of VRUs towards violence reduction.

1.4.2 Assessing VRU progress and wider outcomes

To assess the progress of VRUs towards a whole-systems approach and consider wider VRU outcomes, the team used both a process evaluation and theory-based impact evaluation. The theory-based impact evaluation involved developing and testing contribution hypotheses for each element of the whole-systems approach based on the programme-level ToC. Recognising VRUs are a longer-term approach to prevent violence, and to better understand VRU progress, the team broke the contribution hypotheses down into sequential stages:

• multi-agency working:

  • stage 1: VRUs engage with relevant partners to develop a shared response to violence

Key considerations for this stage included the extent to which: partners were informing the direction of the VRU; a range of partners were (meaningfully) engaging with the VRU; and partner expertise and resources were being utilised by the VRU. Furthermore, although only introduced in January 2023, the team also considered the implications and early effects of the SV Duty within this stage.

• stage 2: VRUs lead and coordinate the shared response to violence

Key considerations here included the extent to which: partners viewed the VRU as leading the multi-agency response to violence; and partners had clear roles in the response to violence, which they were supported (by the VRU) to take ownership of.

• stage 3: A more efficient and effective multi-agency response to violence

This included consideration for the extent to which: the multi-agency approach led by the VRU created efficiencies (for example, reduced duplication); the multi-agency response was perceived as effective; and, aligned to the policy focus for 2022 to 2023, the multi-agency response was considered sustainable. The latter considered the implications of the 3-year funding and match-funding requirements.

• data sharing and analysis:

  • stage 1: VRUs access and analyse data from relevant partners and agencies

Key considerations for this stage included the extent to which: partners were sharing the data (at the required level of granularity and quality to inform the VRUs response); and the VRU was working with partners to improve the quality and granularity of data shared.

• stage 2: VRUs identify and better understand trends in violence and the at-risk cohorts (and hotspots)

Key considerations here included the extent to which: analysis provided the VRU and partners with a better understanding of violence and its drivers; and analysis had identified at-risk cohorts and/or areas.

• stage 3: VRUs’ responses to violence reflect, and are targeted towards, the groups/areas that are most in need

This included considering the extent to which: analysis informed the VRUs’ response (for example, were those identified as at risk subsequently targeted with support) and informed the work/direction of partners; and the focus of the VRUs’ response to violence changed/evolved based on analysis. The former included considering whether the Strategic Needs Assessments (SNAs) developed by VRUs and their Response Strategies aligned.

• young people and community engagement:

  • stage 1: VRUs meaningfully engage young people and communities

Key considerations for this stage included the extent to which and how the VRU was (effectively) engaging with those (most) affected by violence, and whether this was considered meaningful.

• stage 2: As a result of engagement activity, the VRUs’ responses are informed by those of young people and communities

Key considerations here included the extent to which: engagement improved the VRUs’ understanding of the needs of young people and communities; and whether and how this understanding informed the VRUs’ responses. The latter included consideration for interventions co-designed with young people and communities.

• stage 3: VRU activities are more effective at meeting the needs of young people and communities

This included considering the extent to which VRU activities, particularly interventions, were perceived as more effective as a result of the engagement and insights these provided.

• VRU interventions:

  • stage 1: Interventions are strategically commissioned and evidence-based

Key considerations for this stage included the extent to which: VRUs worked with partners to commission interventions strategically; interventions were commissioned based on identified needs and existing evidence of their effectiveness; and VRUs were contributing to the evidence base through evaluation and monitoring. This included an assessment of VRUs focus on interventions defined as ‘high impact’.

• stage 2: Interventions are appropriately targeted and effective at supporting those most in need

Key considerations here included the extent to which: interventions were targeted at, and effectively engaging individuals/areas identified as most in need; and interventions were perceived as effective.

• stage 3: Interventions lead to the intended outcomes for young people and communities

This included consideration for the extent to which: intended immediate/shorter-term outcomes were being realised; and the ultimate/longer-term outcomes of reduced violence were materialising.

The evaluation team drew on a range of evidence to assess the VRUs’ progress within and across stages. The assessment focused on evidence of progress being delivered and whether stakeholders deemed this effective. Relative to previous years’ evaluations, which focused on overall progress since the VRU’s inception, greater focus was placed on progress made in 2022 to 2023. Whilst overall progress was commented on for each element of the whole-system approach, it is important to bear this distinction in mind. As VRUs and their approaches become more established, there is potentially less scope to evidence strides in progress, which are more associated with the inception stage of a programme or policy.

The evaluation team did not include the 2 new VRUs, Cleveland and Humberside, in the progress assessment scoring but did include them in the overall dataset for analysis for this report. These VRUs were primarily focused on setup and early implementation at the time of the research, making their progress incomparable to the other 18 VRUs. Instead, the evaluation team met with these VRUs to discuss and explore their implementation plans.

The team reported progress against each stage at a programme-level (the general progress across all VRUs) using the terms defined in Table 1.1.

Table 1.1: Progress assessment terms

Levels of progress:	Defintion
Good	Broadly consistent evidence of (continued) progress being delivered, and this being perceived as effective
Some	Some evidence of (continued) progress, and this being perceived as effective
Limitied	Limited evidence of (continued) progress, and/or this being perceived as effective

Levels of progress evident in (VRU coverage):	Defintion
All	All 18 VRUs
Almost all/all but a few	15 to 17 VRUs
Most	9 to 14 VRUs
Several	4 to 8 VRUs
Few	1 to 3 VRUs

The process and theory-based impact evaluation included collection and assessment of the following data:

• a desk review including VRU funding applications, delivery plans, quarterly monitoring returns, SNAs and Response Strategies. This included identifying and extracting key information on VRU structures, teams and financing, and details of all funded interventions
• initial interviews with VRU teams (20) to build on the desk review to better understand VRU plans, progress so far, and aims for 2022 to 2023
• interviews with strategic stakeholders in each VRU (78 in total) to understand progress against the ToC, with a particular focus on multi-agency working and data sharing. Strategic stakeholders typically comprised VRU Core Members, who are (senior) representatives from partner organisations. Stakeholders were purposefully sampled to provide coverage of a range of key partners in each VRU
• an online survey of strategic stakeholders to capture a wider (high level) set of views on progress against the ToC. The survey comprised primarily closed questions aligned with the stages of progress (see above). The total number of stakeholders that responded to the survey was 107
• operational stakeholder interviews in each VRU (89 stakeholders in total) to understand progress against the ToC, with a particular focus on more operational elements. The evaluators purposefully sampled stakeholders to capture each VRU structure
• frontline stakeholder interviews in each VRU (65 in total across 18 VRUs as interventions were not operational in Cleveland and Humberside at the time of the research) to understand progress against the ToC, with a particular focus on intervention delivery and outcomes. The evaluators sampled interventions based on intervention typology and target group to provide coverage across a range of interventions. However, they generally avoided interventions subject to (more intense) local evaluations to avoid potential research burden
• young people interviews (with around 20 young people) sought to understand their interactions with the VRU-funded interventions and what difference these were making. Young people were invited to engage with the evaluation through the interventions they were attending but given assurance that their contributions would be confidential – within the limits of safeguarding and need to disclose risk of significant harm. For young people aged 15 and under, evaluators obtained consent from parents/carers and from the young people again on the day
• young people and intervention providers were included in digital ethnography conducted in 3 VRU areas. This took the form of one day’s unstructured on-site observation of the intervention and one day’s engagement with a young person from the intervention, accompanying them in the intervention and through other activities

All interviews were semi-structured and aligned to the programme-level ToC and connected evaluation framework. The evidence generated was used to produce VRU-level case studies. The team analysed all case studies thematically to assess progress against the different stages of the whole-system approach (see above) at a programme-level.

1.5 Limitations

The evaluation was designed to ensure as robust and comprehensive approach as possible, combining multiple methods of data collection and analysis to provide a thorough investigation of how, why and in what context differing levels of success have been achieved within VRUs and at the programme-level.

However, there are some limitations to consider when reading the evaluation findings in this report. Key limitations to the data collection and analysis were:

• the timeframe required for observing outcomes and impacts. The whole-system approach adopted by VRUs was intended to be long term. Longer-term outcomes and impacts, as outlined in the programme-level ToC, are only likely to be fully observable beyond the timescales of this evaluation
• the perceptions of VRU stakeholders primarily informed assessments of progress. Whilst the team interviewed multiple stakeholders, internal (that is, VRU teams) and external to the VRUs (that is, Core Members, community representatives and young people), to gather a range of perspectives to allow for a balanced assessment, it was not always possible to validate these against other sources of data. However, the survey of strategic stakeholders and increased analytical scrutiny of VRU documents and data help mitigate this to an extent
• linked to the above, whilst there was increased scrutiny of VRU documents and data, there were limitations with this. Key examples including not having access to all SNAs and Response Strategies and these now being out of date (new SNAs were in development), and the interventions detailed in delivery plans not always mapping over to quarterly monitoring returns, which restricted analysis
• the scale of research with young people was limited. Whilst this element of the research was not intended to fully represent the VRU programme (which has supported hundreds of interventions and many thousands of young people), it was only possible to engage with young people at one or 2 interventions in a sub-set of VRU sites. These interventions were part of a wider sample of interventions the team approached to provide a cross-section of intervention type. The findings provide important insights into the experiences of young people supported by VRU-funded interventions but do not cover all intervention types, where experiences may differ. Young people were involved from interventions focused on education, sport, and mentoring
• for the QED, a key limitation was variability in measured outcomes, which affects the precision of impact estimates. This was a particular limitation for police recorded crime (PRC). The generalised synthetic approach is designed to minimise variability in the data as far as possible, and variability was less of an issue for the primary outcome measures
• whilst police recorded offending data for Greater Manchester was available (it had not been in previous years owing to a change in systems), it was not included in the QED due to under recording during some periods, which made it unsuitable for assessing changes over time^{[footnote 5]}
• additionally for the QED, it is useful to note that homicides and hospital admissions resulting from violent injury from a sharp object are low-count outcomes. This can make it difficult to detect statistically significant changes

1.6 Report structure

The report contains the following chapters:

• Chapter 2 reports the results from the quasi-experimental impact evaluation (SCM), estimating the impact VRUs have had on reducing SV
• Chapter 3, Chapter 4, Chapter 5 and Chapter 6 provide an assessment of VRUs’ progress and contribution towards each element of the whole-system approach. Each chapter starts with an assessment of progress across stages (see Section 1.4.2) and the evidence of progress within each stage; it then discusses approaches that appeared to be effective (facilitating progress); and concludes with an overall judgement of VRUs’ contribution to the progress made
• Chapter 7 presents conclusions on the extent to which VRUs have contributed to violence reduction and wider aims, recommendations arising, and areas for further investigation

2. Impact on violence

This chapter presents results from the quasi-experimental impact evaluation, which provided quantitative estimates of the impact of SV funding on violence outcomes.

Section 1.4 summarises the methodological approach and outcome focus, and Annex A provides additional methodological detail and outputs from the quasi-experimental impact evaluation.

2.1 Impact on violence outcomes

As discussed in Section 1.4, the precise approach to estimating impacts was generalised synthetic control groups. This was the same approach implemented in previous years’ evaluations. A key consideration for the 2022 to 2023 impact evaluation was the 2 new VRUs, Cleveland and Humberside, moving from the comparator group to the treatment group. To navigate this and provide maximum insights, the following treatment specifications were tested in the synthetic control group analysis:

• Model 1: Cleveland and Humberside initially in the comparison group but then switching to treatment group in 2022 to 2023 financial year. This model aligned to VRU programme initiation dates in each PFA
• Model 2: Cleveland and Humberside in the comparison group only. The justification for this was that the 2 new VRUs were in a setup phase in 2022 to 2023, where direct interventions with young people, and wider activity, that could be expected to impact on violence outcomes were limited. This was consistent with the VRU Impact Feasibility Report, which recommended only testing for impacts on the original 18 VRUs after 18 months of programme delivery to allow sufficient time for the anticipated longer-term preventative effects of VRUs to materialise (Home Office, 2020)
• Model 3: Cleveland and Humberside excluded from the data to test for any potential contamination of estimates resulting from either of the above models

For primary outcome measures, differences in impact estimates between models were limited. As such, and with the above in mind, only results for Model 2 are presented in this chapter; however, Annex A contains the results from all models and discusses the differences between the models.

Table 2.1 details the programme-level cumulative impact estimates for all outcomes of interest following 3 years and 9 months (6 months for homicides) of SV funding (the ‘treatment period’, which covered from April 2019 to December 2022). The team analysed all outcomes as quarterly rates per 100,000 persons using local authority (LA) or broadly equivalent community safety partnerships (CSP) level data. The exception to this was homicides and PFA-level data, which was quarterly rates per one million persons.

In any statistics model, there is some uncertainty around the (impact) estimates due to variability in the measured outcome. Confidence intervals (CI) provided a range of plausible values for the sample mean that were consistent with the observed data. If the CI range included zero (admitting both positive and negative impacts), the team did not consider the impact estimate statistically significant (at the 5% level). Statistically significant estimates have an asterisk in the table.

Table 2.1: Impact estimates

Outcomes	Estimate	Std. Error	Lower CI	Upper CI	P-value
Primary outcomes:
Homicides	-1.53	2.46	-6.35	3.05	0.51
Hospital admissions – Sharp object	-0.9	0.57	-2.14	0.18	0.09
Secondary outcomes:
Hospital admissions – All violence	-7.96*	3.6	-13.99	-0.002	0.05
PRC: Violence with injury	6.89	54.67	-146.84	64.91	0.56
PRC: Violence without injury	-62.95	190.25	-556.57	219.24	0.42
PRC: Possession of weapons offences	-9.45	20.96	-62.5	23.71	0.38

Notes:

1. * indicates statistical significance.

Key observations from Table 2.1 are:

• there were no statistically significant effects on the primary outcome measures of homicides and sharp object hospital admissions
• the impact estimate for homicides was a reduction of 1.5 (per one million persons) but this was not statistically significant. Whilst this central estimate is less of a reduction than that reported in the 2021 to 2022 evaluation (-2.84, 95% CI [-6.67, 1.05]), the CIs are overlapping so we cannot say the results are (statistically significantly) different. It is also worth noting that homicides is a low-count outcome, which can make it difficult to detect statistically significant effects. Power analysis was conducted and indicated a larger effect (a change of 5 per one million persons) would be detectable (at the 80% power level)
• the impact estimate for hospital admissions resulting from violent injury with a sharp object was a reduction of 0.9 (per 100,000 persons) but this was not statistically significant. Similar to homicides, sharp object hospital admissions are a lower-count outcome. However, as indicated by the CIs (and supported by power analysis), a slightly larger reduction would have likely been detectable as statistically significant
• there was a statistically significant reduction of -7.96 (per 100,000 persons) for the secondary outcome of hospital admissions resulting from any violent injury
• results for homicides and hospital admission outcomes were similar across the different model specifications

For PRC outcomes, there were no statistically significant effects. The central estimates indicated smaller and not statistically significant effects compared to previous years. However, the wide CIs for these estimates, which indicate a high degree of variability in the data, overlapped with previous years. As such, we cannot say that the results for this outcome reflect a real change in violence. The findings this year are nonetheless encouraging, showing impact on more serious levels of violence compared to previous years, and consistent with previous research showing that impact on the most SV may take longer to materialise.

The following graphs and discussion provide additional outputs from the synthetic control group analysis for the primary outcomes and hospital admissions resulting from any violent injury (secondary outcome where there was a statistically significant reduction). Outputs for other outcomes are provided in Annex A, Note that all charts report in financial quarters and the financial year ending.

Figure 2.1 shows the average quarterly hospital admissions for sharp object violent injury rates (per 100,000 persons) for VRU areas and the synthetic control group constructed. Relative to the (unweighted) average hospital admissions for sharp object violent injury rates in non-VRU areas, the synthetic control group provides a better counterfactual. The increase in quarter 2, 2021 (July to September) corresponds to restrictions lifting following the first national COVID-19 lockdown. Differences following SV funding in hospital admissions for sharp object violent injury rates between VRU areas and the synthetic control group (which would indicate an impact) initially appear limited but trends in VRU areas appear to track below the synthetic control group in more recent quarters.

Figure 2.1: Hospital admission rates for sharp object violence (fiscal quarters) – synthetic control group analysis

Figure 2.2 shows the average quarterly homicide rate (per one million persons) for VRU areas and the synthetic control group constructed. Whilst relative to the (unweighted) average homicide rates in non-VRU areas, the synthetic control group provides a better counterfactual; there was some divergence in trends in 2018. During the financial years ending 2021 and 2022, the trends in VRU areas tended to track below the synthetic control group (indicating a potential impact) but this was not sustained in more recent quarters.

Figure 2.2: Homicides (fiscal quarters) – synthetic control group analysis

Figure 2.3 shows the average quarterly hospital admissions for any violent injury rates (per 100,000 persons) for VRU areas and the synthetic control group constructed. Relative to the (unweighted) average hospital admissions for any violent injury rates in non-VRU areas, the synthetic control group provides a better counterfactual. Following SV funding, trends in VRU areas tended to track below, or very close to, those in the synthetic control group. Differences were most apparent in the most recent quarters.

Figure 2.3: Hospital admission rates for any violence (fiscal quarters) – synthetic control group analysis

In previous years’ evaluation reports where there have been statistically significant reductions in outcome measures, cost-benefit analysis was undertaken. Key to enabling this analysis was available and reliable estimates of the estimated cost savings associated with specific outcome measures. Following discussions with the Home Office, no such estimates were currently available for the outcome of hospital admissions resulting from any violent injury. Whilst consideration was given to the costs savings associated with police recorded violence with injury offences, this was ultimately rejected as they likely underestimate the cost of those injuries that result in a hospital admission. Further research is required and will be pursued by the evaluation team to estimate cost savings for potential inclusion in future evaluation reports.

2.2 Hospot analysis

To understand the impact of SV funding in areas most affected by violence, the evaluation team undertook exploratory synthetic control group analysis using Lower Super Output Area (LSOAs)-level police recorded violence data. The areas most affected by violence (‘hot spots’) were defined as LSOAs with the highest levels (top 5% within each PFA) of violence prior to SV funding. Recognising the analysis was exploratory, and that the analysis was limited to just 8 SV-funded areas and 8 non-SV-funded areas where LSOA information was recorded well and consistently over time^{[footnote 6]}, we advise some caution with the results from this analysis. It is also important to note the aforementioned limitations of police recorded data (see Section 1.5) and that the number and composition of SV-funded areas included in this year is different to those analysed in previous years.

The analysis was intended to primarily provide insights into the impact of Grip (and its predecessor, Surge), where there was an explicit focus on hot spot policing. Whilst VRUs were expected to target activity at individuals and areas most in need, it was anticipated that their impact would be PFA-wide.

The estimated total (statistically significant) reductions (over 3 years and 9 months), relative to the synthetic control group, were:

• for violence with injury offences, -17,424 (95% CI [-21,182, -13,920])
• for violence without injury offences, -9,351 (95% CI [-17,563, -1,564])

These results provide additional insights into the potential impacts of SV funding. Whilst the programme-level analysis (Table 2.1) did not detect a statistically significant effect on these outcomes, the analysis of hot spots indicates a reduction in violence in the most affected areas.

2.3 Exploring differences in impact and approaches

The evaluation team undertook meta-analysis to determine whether there was any significant variation in the impact seen across the VRUs. They then used meta-regression to explore whether different levels of SV funding or approaches on how the funding was used could explain differences in impacts between VRUs. This used the impact estimates generated using synthetic controls for each VRU and then treating these as individual studies. Analysis focused on the impact estimates for hospital admissions outcomes as these are not prone to the limitations (for example, differences in recording) of police recorded outcomes, and the low counts associated with homicide outcomes. Based on the recommendations from the previous year’s meta-regression, the team tested explanatory variables relating to VRU funding levels or focus.

Significant variation between VRU-level impact estimates were detected for the effect on hospital admissions, both from sharp objects and any violence. The following models were tested for each outcome; Table 2.4 shows which showed a statistically significant (at the 95% level) result.

Table 2.4: Models that returned a statistically significant result for different types of hospital admissions

Explanatory variable	Hospital admissions from sharp object	Hospital admissions from any violence
Funding per capita (£ per person)		X
Spending on interventions per capita (£ per person)	X	X
Spending on high-impact interventions	X

This shows that impact estimates for VRUs with higher spending on interventions per capita were more likely to indicate a reduction in hospital admission from both sharp objects and any violence. VRUs with more funding (per capita) were also more likely to indicate a reduction in hospital admissions from any violence, and VRUs that spent more on high-impact interventions were more likely to show a reduction in hospital admissions from a sharp object. It should be noted that 2 VRUs were flagged as outliers to this pattern. When these VRUs were removed from the analysis, there was no longer a statistically significant result across any models, but the trends were still apparent.

2.4. VRU stakeholder perspectives of imapct

This section considers wider evidence and perspectives of the impact of VRUs.

Figure 2.4 shows results from the Core Member Survey in response to the question “Which of the following statements best describes your view on the impact of your VRU on reducing violence?” Almost half (46%, n = 49) responded that they thought the VRU was helping to reduce the violence and that this was evident in local data and/or intelligence; whilst 39% (n = 42) of respondents thought the VRU might be helping to reduce violence but evidence of this was less clear.

Figure 2.4: Core Member Survey – Perceptions of VRU impact on reducing violence

Core Members were also asked if they thought VRUs were the right approach to reduce violence. A majority (58%, n = 62) responded yes and 38% (n = 41) said potentially, but changes are required. Those that responded yes elaborated that they appreciated the collaborative, multi-agency nature of VRUs, as well as their focus on early intervention, prevention and targeted local strategies. In terms of changes, common themes included:

• increasing funding for youth services, early years and education to address the root causes of violence
• improving collaboration with voluntary and community sector organisations (VCSOs) for more effective interventions
• further enhancing coordination, communications and transparency across stakeholders
• securing long-term funding to enable better commissioning of evidence-based interventions, staff retention and, more widely, sustainability
• shifting the focus from police-led Units to LA-led for a more place-based response

3. Multi-agency working

This chapter provides an assessment of VRUs’ progress and contribution towards effective multi-agency working. Multi-agency working is a core element of the whole-systems approach to violence prevention and can be considered a foundation for other elements.

Aligned to the programme-level ToC, progress towards effective multi-agency working is considered against the following connected and sequential stages:

1. VRUs engage with relevant partners to lead the development of a shared response to violence
2. VRUs lead and coordinate the response to violence, where ownership is shared with partners, which ultimately leads to
3. a more efficient and effective multi-agency approach

The assessment was informed through the triangulation of interviews with multiple stakeholders, a review of supporting documentation, and responses to the Core Members survey. As mentioned previously, there is a particular focus on progress made in 2022 to 2023 and the policy priorities related to multi-agency working. These priorities included the introduction of the SV Duty, 3-year funding allocations, and the sustainability of VRUs.

The chapter follows the progress assessment with discussions of VRU approaches to multi-agency working, which facilitated progress.

3.1 Progress towards multi-agency working

Overall, VRUs have continued to make good progress towards embedding an effective multi-agency approach. This included progress in engaging a wider range of partners, utilising partner expertise and resources, and coordinating multi-agency responses to violence. The introduction of the SV Duty and 3-year funding were generally viewed as an opportunity to strengthen partner buy-in, commitment and focus, but there was a recognition it would take time to realise the full impacts of these. There were still areas for improvement noted in terms of meaningful engagement and shared responsibilities with all partners, and ensuring the sustainability of the multi-agency approach.

Common themes of progress in 2022 to 2023 included:

• improved engagement and collaboration among partners, although this can be uneven. Challenges persisted with the engagement of some partners, particularly health and education
• evolving understanding of partner roles and, linked to this, more effective utilisation of and shared responsibilities with partners, with some variation noted
• stronger multi-agency working, leading to further reduced duplication, joint working and culture change reported in many VRUs
• indication of the SV Duty and 3-year funding contributing (or having the potential) to increased focus, engagement and resource commitment in many VRUs

In terms of VRUs’ overall approach to multi-agency working, since their inception, stakeholders interviewed reported these were generally effective. The Core Member Survey mirrored this view, with the majority (around 90%) reporting different elements of the approach were either highly or somewhat effective. Whilst most VRUs had broadly kept the same approach and model developed in previous years, others had undergone substantial change following a review (see Section 3.2.1), and some have increased their local focus. The former caused some inevitable disruption but was strongly viewed by stakeholders as a positive change.

Progress against each stage of multi-agency working is discussed in the sections that follow. Evidence relating to progress was generally strong (at least broadly consistent within stakeholder groups).

3.1.1 Development of shared response to violence

The Home Office expected all VRUs to engage with relevant partners to develop a shared response to violence, which draws on partner expertise and resources. This engagement ensures partners buy into the shared response to violence. An early anticipated effect of the SV Duty was the required engagement of key partners to work together to reduce violence, which VRUs can provide the structure and forum for.

Overall progress in 2022 to 2023 against this stage was good in most VRUs. Progress tended be more limited where levels of partner engagement were more variable. In terms of overall perceived effectiveness, more than half of Core Members responding to the survey (52%, n = 56) thought that engagement with the VRU was very effective, and a further 40% (n = 43) reported it was somewhat effective. The small proportion responding that the engagement was somewhat ineffective or were indifferent (8%, n = 8) were concentrated in a few VRUs and were typically Voluntary, Community and Social Enterprise (VCSE) or LA representatives^{[footnote 7]}. This pattern might reflect wider views of VCSEs being relatively less engaged by the VRU, and VRUs having/aiming to engage multiple LAs. Results were similarly positive for responses to the question relating to the utilisation of partner expertise and resources, with 43% (n = 46) responding this was very effective and 46% (n = 49) responding somewhat effective.

Generally, VRUs maintained and improved on their engagement with partners to inform their strategic direction. This was most evident in VRUs where stakeholders reported changes being made to the Response Strategy based on their engagement, and where VRUs had brought in new roles or structures to help expand and strengthen engagement with partners. However, there was some variation across and within VRUs with a feeling from stakeholders that whilst they had been consulted, this was not necessarily driving the direction of the VRU, and that the direction was more driven by the police.

Progress in meaningful ongoing engagement with and between partners was noted in all VRUs. This was typically associated with regular and appropriately focused and attended strategic meetings. However, engagement did vary based on several factors including competing organisational priorities and different perspectives on how to prevent violence, between partners. Some VRUs noted continued issues with engagement with health and education partners.

Most VRUs demonstrated progress in effectively utilising partner expertise and resources in their response to violence. Aligned to progress with general engagement, progress was most evident in VRUs that had developed roles and structures aimed at better utilising their expertise and resources. In some VRUs, this included more place-based work to develop and harness expertise at a more local level. Variation typically resided within VRUs with some partners being less engaged than others and, as such, offering less in terms of expertise and resources.

There was evidence of the 3-year funding and SV Duty supporting progress in engagement with partners (see Section 3.2). However, the impact of the SV Duty was more of an anticipated effect.

3.1.2 VRUs lead and coordinate the response to violence

The core function of VRUs was to lead, coordinate and support partners to share ownership of the local response to violence. The Home Office considered a shared response to violence, which harnesses the resource and expertise of partners, key to VRUs achieving their ultimate aim to prevent violence.

Overall progress in 2022 to 2023 against this stage was broadly split between ‘some’ and ‘good’ across VRUs. Progress was most evident where partner roles had been strengthened and clarified, and partners were taking ownership of these roles. Again, the extent of variability in partner engagement tended to be a driver for progress made. In terms of overall effectiveness, 44% (n = 47) of those responding to the Core Member Survey thought that the VRUs leadership and coordination of the response to violence was very effective. A further 47% (n = 50) thought the approach was somewhat effective. Responses were similar for a question covering the support provided to partners by the VRU.

It was evident in most VRUs that they were viewed as leading or coordinating the multi-agency response to violence. Partners felt supported and involved in the response to violence, and they highlighted the VRUs for their mature partnership approach, growing clarity in roles and priorities, coordination and shared ownership of SV (see below). The VRUs were also recognised for their expertise, data analysis, evaluation capabilities and strong leadership. In the few VRUs where this was less evident, reasons included overlap and complexity within their area, variable stakeholder perspectives, and variance at the ‘hub’ level (in hub and spoke models – see Structures to enable multi-agency working in Section 3.2.1 for further detail on VRU models).

There were clear roles for partners in most VRU responses to violence, which was appreciated, but this view was more mixed in a few VRUs. Reasons for the more mixed views included gaps in strategy implementation, differences in stakeholder perspectives, and overlap with existing initiatives. In terms of partners taking ownership of their roles, this was again evidenced in most VRUs and facilitated by support from the VRU and collaboration among agencies. However, some VRUs experienced challenges related to organisational priorities and varying levels of engagement among partners. There were mixed levels of accountability across VRUs, with some having clear mechanisms in place and others facing challenges in ensuring partners delivered on their roles. It was too early to say whether the SV Duty would help improve partners ownership and accountability, but there was a general view that it had the potential to, particularly for those partners that have been historically less engaged.

It was evident across VRUs that support was being provided to partners to help them deliver on their roles within the response to violence. The types of support provided varied among the VRUs, but most commonly included funding, capacity building, trauma-informed training, and sharing of information and evidence. In some cases, the VRU provided tools such as evaluation toolkits, best practice guidance, and informative webinars to help partners understand the drivers and root causes of SV. There was variability across VRUs, with some offering fewer instances of more targeted support, and others offering a broader range of activity.

3.1.3 A more efficient and effective multi-agency approach

As a result of the preceding stages, it was anticipated there would be a more efficient, effective and sustainable multi-agency approach towards violence prevention. In previous years’ evaluations, this stage (which includes organisational and professional culture change) was considered a longer-term aim, and potentially too early to measure. Recognising VRUs were in the fourth year of operation in 2022 to 2023, it was reasonable to expect progress, even if these are still early signs.

Overall progress in 2022 to 2023 against this stage was broadly split between ‘some’ and ‘good’ across VRUs. Progress was most evident where, due to the VRU’s work, cooperation and collaboration between partners had improved, which enabled more efficient and effective response to violence. Variability in progress made related to levels of partner engagement, clarity of roles and the support provided by the VRU to partners.

Enhanced partner capacity was noted in some VRUs, where the support provided by VRUs, such as trauma-informed strategy implementation, education leads focused on reducing exclusions and youth work leads, was benefiting partners. Particular efforts to reduce duplication were also evident in some VRUs. These included complementing the work of partners and refining referral pathways and criteria.

Whilst sustainability generally remains a concern for VRUs, the 3-year funding was highly appreciated and has allowed for longer-term planning, including the mainstreaming of projects (with partners) and minimising the reliance on Home Office funding for core team positions. Alongside the SV Duty, the 3-year funding has also supported the engagement and commitment from partners, particularly for those where the historical short-term funding had proved a barrier for them.

3.2 VRU approaches to multi-agency working

The common themes for progress towards effective multi-agency working in 2022 to 2023 included strong relationships with partners, effective structures, strong leadership and having a clear vision. Key effective approaches within these themes include strategic appointments, improved governance structures, data sharing, long-term funding and alignment with broader strategies.

Each theme is discussed in turn in the next section and followed by a discussion of common challenges. The approaches and challenges detailed were those most prominent in the evidence collected in 2022 to 2023, but should be considered against findings in previous evaluation reports.

3.2.1 Effective approaches to multi-agency working

Building and embedding strong working relationships with partners

Building and embedding strong working relationships among partners was a key focus across most VRUs. Key to this was a people-centred approach to relationship building, increased communication, support and multi-agency collaboration. This included appointing individuals, or utilising those already engaged with the VRU, who have pre-existing relationships and shared values with partners. A notable example included changes in policing personnel to include those with values better aligned with other VRU stakeholders, which supported better collaboration and a more effective response to violence.

Longer-term funding and strategic planning enabled better engagement and commitment from partners. There were examples of partners working together to develop and implement multi-year interventions, which helped develop relationships and buy-in, and would not have likely happened in the absence of the 3-year funding.

There was also evidence of VRUs starting to harness the SV Duty to increase partner engagement and ownership in addressing SV. The SV Duty has helped change the narrative with local partners, as they now had statutory responsibility to work on the SV agenda, which has driven engagement with the VRUs as a result.

Structures to enable multi-agency working

Effective governance structures and local partnerships contributed to progress. Governance structures, such as local partnership structures and embedding local VRU coordinator roles, helped in building cooperation. In one VRU, the establishment of an Executive Board overseeing the Steering group and various working groups for the spokes (see below) improved accountability and coordination.

Whilst stakeholders in most VRUs were unable to discuss the merits of different VRU models, as they have typically only been exposed to one, 2 VRUs had undergone a model change in 2022 to 2023. For context, we can summarise different VRU models as:

• centralised: All VRU activity and funding managed centrally from a central Unit
• hybrid: VRU managed centrally from one Unit but either devolves a proportion of funding to LA areas (via existing structures/organisations, such as CSPs) and/or has ‘locality leads’ within the VRU to commission and deliver more localised activity
• hub and spoke: VRU activity overseen from a central Unit (the ‘hub’) but funding devolved to/responsibility given to specially created sub-Units (the ‘spokes’) to commission and deliver local activity

The first VRU has undergone a substantial change by moving from a hub and spoke model to a hybrid model. This was based on recommendations from the Home Office and others. Whilst implementing the change limited the progress they could make across the different elements of the whole-systems approach this year, all strategic and operational stakeholders interviewed strongly believed changing the model was the right decision. Stakeholders thought that the hybrid model was laying the groundwork for a more efficient and effective multi-agency response. Partners are more engaged and clearer on their roles, and having a more central approach has allowed for the development of 3 large-scale interventions/projects, which would not have been possible under the hub and spoke model.

The other VRU has moved from a central approach to a more hub and spoke/hybrid model. Nine place-based partnerships (spokes of the hub and spoke model) and VRU coordinator roles were introduced to embed local work and report back to the operational and strategic Boards, which supported partner engagement and buy-in at a more local level. The driver for this was the anticipation of the SV Duty, which is more focused on partnerships at a LA level. This move has strengthened and made decision-making more inclusive by empowering local areas whilst maintaining a central point of effective strategic coordination. The latter includes interventions being made available across areas, rather than many separate interventions in each area.

Whilst it is difficult to say which model is best, with some VRUs arguing that a different model would not work in their area, the above and wider considerations around the importance placed on local partnerships within the SV Duty, and sustainability suggest the hybrid model has an advantage. Under this model, VRU teams can be kept relatively small by utilising existing partnerships and structures at a more local level, whilst maintaining central decision-making for effective coordination of the multi-agency response to violence.

The VRUs’ function and value in data sharing and intelligence gathering was an enabler for engaging partners and motivating action. This was demonstrated through dashboards and data-driven approaches to understand the drivers of SV and informing the targeting of interventions. One VRU introduced the Injury Surveillance Dashboard, which helped to further assure buy-in from health partners and reinforce a data-driven response to understanding SV drivers.

Leadership

Strong leadership was noted as a key enabler for effective multi-agency working in some VRUs. Leaders with credibility and varied backgrounds in the field fostered trust among partner agencies. In one VRU, the appointment of a public health specialist within the leadership team facilitated a strategic approach to commissioning interventions and broadened relationships beyond crime and community safety networks.

For another VRU, the head of the VRU sitting on the CSP board led to increased integration with partners and improved communication. This showcased the importance of having leaders who actively participated in strategic planning and decision-making with partner organisations.

Vision

A clear vision and buy-in from key operational and strategic partners was important. Embedding a public health approach through events, communications and data work were considered effective. This was closely linked to the importance of building relationships with partners.

For some VRUs, alignment with broader government strategies, such as the Beating Crime Plan, national police and crime measures, and drug strategies, influenced priorities and contributed to the sustainability of the multi-agency approach. For example, in one VRU, partners worked together with the VRU on a joint bid for funding, which not only demonstrated efforts towards sustainability, but the collaboration brought partners together around a shared vision and aims.

Additional points noted included the importance of engaging a wide range of partners to understand and consider their vision and priorities. In some VRUs, there was evidence of this having wider benefits to the VRU, such as facilitating better intelligence gathering and data sharing. Although only explicitly noted in one VRU, the establishment of a subgroup of strategic stakeholders focused on developing action plans to operationalise the strategic vision was considered effective.

3.2.2 Challenges to effective multi-agency working

Particularly pertinent key challenges, or not just the inverse of the enablers already discussed, were noted by several VRUs, including:

• securing meaningful engagement from all partners, particularly from health and education sectors, remained a challenge. Competing priorities within organisations, resource constraints, and the need for relationship building can slow down progress. Some VRUs also faced difficulties in engaging meaningfully with partners, which might lead to other partners perceiving the Unit as too police focused (as they are most engaged)
• whilst the 3-year funding was broadly welcomed, some still flagged the uncertainty around funding (having to reapply to the Home Office each year and/or what happens after the 3 years) and were nervous about sustainability, which impacts their ability to engage partners and plan for the long term
• local and organisational politics can interfere with VRU delivery and create conflicts in discussions. Overlapping responsibilities and complexities in legislation must be navigated locally. Political dynamics, such as those between LAs, can also hinder engagement and collaboration
• sustainable approaches to data sharing with partner organisations can be a challenge, especially for SNAs and understanding SV trends across different agencies. This is further complicated by varying geographical boundaries in datasets

3.3 VRU contribution to multi-agency working

VRUs’ contribution to multi-agency working was significant in around half of the VRUs, where it was unlikely this progress would have been made without the VRU. In other VRUs, the contribution had accelerated/facilitated progress, which may have been made in the absence of the VRU over a longer period/at a reduced rate. Where VRUs were changing their structures or models, there was uncertainty around the VRUs’ contribution.

This assessment of VRUs’ contributions to multi-agency working were mirrored in responses to the Core Member Survey. When asked how they would best describe the VRU’s contribution to multi-agency working locally over the past year, 37% (n = 40) said it was a lot more efficient and effective, and 48% (n = 50) said it was somewhat more efficient and effective. Other responses were spread across several VRUs and tended to come from LA and VCSE representatives.

Figure 3.1: Core Member Survey – perceptions of VRU contribution to multi-agency working

VRUs’ main contribution was their focus on leadership and coordination, bringing partners together with a common aim and supporting evidence-led delivery. Specific examples of VRUs’ contribution was noted for their work in areas where they are responsible for the SV Duty, developing LA violence reduction and vulnerability action plans, and supporting the sharing of best practice.

In some VRUs with existing multi-agency structures, most stakeholders considered the focus and additional resource of the VRU a valuable contribution. Yet, some partners felt VRUs imposed on their existing multi-agency work. Other stakeholders suggested the framework for responding to violence already existed (as well as the drive to address it politically) arguing that related work would have proceeded regardless.

However, some delivery and sustainability planning continued to be influenced by local level priorities, which was supported where VRUs had local partnership structures that helped to build links and address competing priorities.

4. Data sharing and analysis

This chapter provides an assessment of VRUs’ progress and contribution towards effective data sharing and analysis. Data sharing and analysis is a core element of the whole-systems approach to violence prevention.

Aligned to the programme-level ToC, we consider progress towards effective data sharing and analysis against the following connected and sequential stages:

1. VRUs access and analyse data from relevant partners and agencies
2. VRUs better understand trends in violence and the at-risk cohorts and areas; which ultimately leads to
3. a response to violence that reflects and is targeted at those in need

The assessment was informed through the triangulation of interviews with multiple stakeholders; a review of quarterly returns, SNAs and Response Strategies; and responses to the Core Members survey. As mentioned previously, there is a particular focus on progress made in 2022 to 2023 and the policy priorities related to data sharing and analysis. These priorities focused on improving the granularity and quality of data accessed.

The progress assessment is followed with discussions of VRU approaches to data sharing and analysis, which facilitated progress.

4.1 Progress towards data sharing and analysis

Overall, VRUs have made some progress towards effective data sharing and analysis in 2022 to 2023. Progress was most evident in VRUs that had (continued to and had further scope to) invest in systems and access more and better quality data to improve the VRUs’ and partners’ understanding of those at risk. However, there was a wide range of variability across VRUs. This included many still facing challenges including access to granular data, ongoing efforts to increase data sharing and standardise data capture, and organisational restructures hindering data sharing progress.

Common themes of progress in 2022 to 2023 included:

• for many VRUs, improved data sharing and collaboration between various partners, such as police, LAs and health; although this was not the case for other VRUs, particularly with health data
• development of data dashboards or analytics tools in an increasing number of VRUs to better understand trends and hotspots, and inform decision-making
• increased capacity for data analysis (and evaluation) in several VRUs, such as hiring new staff (including but not limited to monitoring and evaluation leads), integrating teams or utilising external expertise

In terms of VRUs’ overall approach to data sharing and analysis, since their inception, views were mixed. The responses to the Core Member Survey highlighted this, with greater proportions saying elements of the approach were only somewhat effective or indifferent, relative to the responses related to multi-agency working.

We discuss progress against each stage of data sharing and analysis in the sections that follow. Evidence relating to progress was generally strong (at least broadly consistent within stakeholder groups).

4.1.1 VRUs access and analyse data

The Home Office expected all VRUs to work with partners to access and analyse a range of data. The drivers of violence and risk factors are complex, and data related to them spread across different agencies. VRUs have a key role in bringing this data together. There was an increased policy focus in 2022 to 2023 for VRUs to access more granular data to support the identification and targeting of at-risk groups and areas.

Overall progress against this stage in 2022 to 2023 was mixed, with around half of VRUs showing good progress and others just some or limited progress. Variation in progress was typically associated with (lack of) access to specific datasets, such as health, and improvements in data quality. Responses to the Core Member Survey reflected these points, with only 29% (n = 31) responding that VRUs data access was very effective. Most (54%, n = 58) thought it was somewhat effective and 15% (n = 16) were indifferent or responded somewhat ineffective. When asked about support from the VRU to improve the quality and granularity of data, Core Members were less positive, with fewer (17%, n = 18) responding very effective and over a fifth (22%, n = 23) indifferent or somewhat ineffective.

In terms of VRUs accessing data, there were ongoing issues for many in accessing (granular, individual case-level) health data. There was some evidence of additional data being accessed in some VRUs but in others there had been little change from the previous year. Some VRUs mentioned the SV Duty might help unblock challenges.

Table 4.1 shows the number and proportion of VRUs accessing and using at least one source of different data types. All VRUs where final quarterly monitoring returns were available (17) were using crime data. And over half were using hospital admissions (71%, n = 12), Education and social care (both 59%, n = 10) and ambulance (53%, n = 9) data. Accident and emergency (A&E)/Information Sharing to Tackle Violence (ISTV) data (NHS Digital, 2014), which is considered high value to understanding violence, was only being accessed and used by 5 VRUs.

Table 4.1: Data types by VRU

Data type	Number of VRUs using	%
Crime	17	100%
Hospital admissions	12	71%
Education	10	59%
Social care	10	59%
Ambulance	9	53%
Housing	6	35%
Other	6	35%
A&E/ISTV	5	29%
Health - other	5	29%
Youth offending	5	29%
Socio-demographic	4	24%
Fire	3	18%

Source: Analysis of VRU Quarter 4 Monitoring Returns

Notes:

1. Returns missing for 3 VRUs.

Many VRUs showed progress in improving the quality and granularity of data accessed. This was evident where VRUs had explored additional datasets, worked with external academic partnerships and increased internal expertise. Other VRUs had made efforts to improve data quality, through initiatives like multi-agency data groups, but were still facing challenges. Challenges included concerns about individual-level data access, proportionality, and inconsistent data sharing among partners.

Table 4.2 shows the number of datasets being accessed and used by VRUs by data type and level of granularity. Except for crime (10), VRUs generally had limited access to Level 3 data, which is individual-level data. Most data was accessed at Level 2 (sub-areas) or Level 1 (VRU-area level). Level 2 data enabled VRUs to identify hotspots.

Table 4.2: Data type by level of access

Data type	Level one (VRU-area level data)	Level 2 (sub-area level data)	Level 3 (individual-level data)
A&E/ISTV		5
Ambulance		9
Crime	3	6	10
Education	3	5	4
Fire		3
Health – other	3	2
Hospital admissions	5	5	2
Housing	2	3	3
Other	2	3	1
Social care	2	5	4
Socio-demographic	3	1
Youth offending	2	2	1

Source: Analysis of VRU Quarter 4 Monitoring Returns

Notes:

1. Returns missing for 3 VRUs.
2. Those accessing Level 3 data likely have access to Level 2 and Level one data too, but this was not necessarily recorded.

4.1.2 Improved understanding of violence and the at-risk cohorts

A key role of the VRU is to undertake analysis to develop a better understanding of violence trends and its drivers and identify the at-risk cohorts (and/or areas) to target support towards.

Overall progress against this stage in 2022 to 2023 ranged between some and good for the majority of VRUs. Progress was most evident in VRUs where systems and analysis were becoming more advanced, and there was an appreciation of the data and insights generated by partners. Core Members surveyed were generally positive about the data shared by VRUs to inform their understanding of violence and the at-risk cohorts, with 36% (n = 39) responding very effective and 50% (53) responding somewhat effective.

Progress towards analysis providing the VRUs and partners with a better understanding of violence and its drivers was good in most VRUs. This included the development and use of SNAs, profiles for sub-areas, and regular updates and knowledge sharing. Progress in other VRUs was more limited, with issues including concerns about overreliance on police data and insufficient granularity of data. It was also noted in one VRU that although analysis had not substantially changed the understanding of violence, it had validated existing knowledge.

There was a similar pattern for progress towards, and reasons for variation in, VRUs’ analysis identifying at-risk cohorts and areas (meaning, ‘hotspots’). Progress was most evident in VRUs that had deployed data dashboards or similar tools. However, there were some concerns there was too much of a focus on identifying hotspots, which could lead to decision-makers concentrating efforts in particular areas when there are wider factors to be considered, such as individual and community needs.

4.1.3 Responses to violence reflects and targets those in need

Resulting from the preceding stages, the VRUs’ and partners’ response to violence should reflect the needs of and target those identified as at risk. This is key for an evidence-based approach where resources are targeted to have maximum effect.

Overall progress against this stage in 2022 to 2023 was mixed, with some VRUs showing good progress and others just some or limited progress.

In terms of the analysis conducted informing the VRUs’ response to violence, stakeholders interviewed considered this to be good in most VRUs. This was evident in analysis helping to identify hotspots, at-risk individuals, and gaps in services, which lead to targeted interventions and collaboration amongst partners (see Section 3.2.1). Variation in progress was typically explained by factors such as the effectiveness of data sharing and analysis, and staff changes. The VRU’s analysis also influenced the partners’ response. Examples included joint action, targeted interventions, enhanced referral pathways, increasing wardens in city centres and informing the police response.

To explore this further, all available VRU SNAs and Response Strategies (and delivery plans) were assessed for alignment. In summary, direct and specific alignment between the needs identified and VRU activities was limited and difficult to assess. However, it is important to note that these documents were not available for all VRUs, some were drafts, and all VRUs were in the process of developing new SNAs and Response Strategies, with instruction from the Home Office to ensure better alignment. Furthermore, all VRUs conduct additional and ongoing analysis, beyond that covered in the SNAs, which may explain the differences in stakeholder perspectives (above) and this assessment. Key points from the assessment included:

• Strategic Needs Assessments:
  • significant overlap across VRUs, with similar theories and data sources used
  • identification of specific local hotspots was rare
  • similar demographics for perpetrators and victims across VRUs
  • holistic focus on early prevention and risk factors, and less emphasis on specific problem areas
  • lacked clear conclusions
• Response Strategies focused on general ways of working and collaboration.
• alignment between SNAs, Response Strategies and delivery plans:
  • little clear alignment between SNAs, Response Strategies and delivery plans
  • interventions mostly based on general theory and similar across VRUs
  • interventions not usually limited to specific demographic groups but target individuals with certain characteristics
  • most interventions occur across the whole PFA, so difficult to assess if funds are invested in hotspots

Assessing alignment between SNAs, Response Strategies and interventions was difficult because SNAs provided a broad overview, whilst Response Strategies focused on general ways of working and collaboration. The logical narrative connecting these documents was often missing, which made it challenging to understand why certain priorities were chosen over others. Interventions were similar across VRUs and appeared to be based on general theory, rather than directly addressing the specific needs identified in SNAs. Furthermore, the whole-system and public health approach encouraged a holistic focus, which made it difficult to pinpoint specific priorities and assess alignment with interventions.

4.2 VRU approaches to data sharing and analysis

The most common themes for progress towards effective data sharing and analysis in 2022 to 2023 include strong relationships with partners, data governance and technological solutions, and the embedding of analysts and analytical capabilities.

The following section discusses each theme, followed by a discussion of common challenges. The approaches and challenges detailed were those most prominent in the evidence collected in 2022 to 2023 but should be considered against findings in previous evaluation reports.

4.2.1 Effective approaches to data sharing and analysis

Whilst there was variation across VRUs, common themes associated with effective data sharing and analysis are detailed below.

Role of key staff and relationships

Linked to effective multi-agency working, key personnel in partner organisations able and willing to assist with data sharing was a significant factor contributing to the progress of VRUs. Over time, the development of positive relationships and collaborations had improved information sharing among partners.

Data governance and technological solutions

VRUs achieved improvements in data sharing through a focus on data governance and the adoption of innovative technological solutions, such as commissioning external suppliers to deliver dashboard approaches or utilising software such as Microsoft Power BI for automation of data queries and reports. The development and use of dashboards provided valuable insights, showcasing the potential of a data-driven approach and reinforced partners’ willingness to share data. The automation of analysis allowed the analysts and VRUs to spend more time on interpretation.

Embedded analysts and analytical capabilities

Progress was associated with the presence of embedded analysts, who could produce the required analysis and work with partner organisations to access data, and the enhancement of analytical capabilities through collaboration.

Funding and stability

Although only mentioned by a few VRUs, sufficient funding and long-term stability enabled VRUs to secure commitment from partners and strengthen the value derived from data sharing. The 3-year funding was specifically noted as an enabler in this sense.

Co-location and direct access to data

Co-location with partner organisations, such as the police, facilitated efficient data flow. Partners demonstrated enthusiasm for data sharing, recognising the value it added to their work. Direct access to data under legal frameworks, such as the Crime and Disorder Act, also played a role in effective data sharing. There was an anticipation in some VRUs that the SV Duty would help overcome some of the persistent challenges with some partners’ data sharing.

Evaluation and best practices sharing

The evaluation of work done by VRUs helped to identify areas needing attention and engagement. An example was provided of the more qualitative evidence collected through evaluations being able to supplement and fill the gaps of more quantitative data and monitoring. Coordinating meetings to share best practices among partners across the VRU area and sharing top-level data with partners and intervention providers contributed to the targeting of resources and the overall progress of VRUs.

4.2.2 Challenges to data sharing and analysis

Common challenges were typically the inverse of enablers and many have been discussed throughout this chapter. In summary, the key challenges faced by VRUs for data sharing and analysis included:

• difficulty accessing health, probation and individual-level school data due to legal restrictions, geographical limitations and anonymisation issues
• inconsistencies in data quality and granularity, as well as challenges in standardising data collection and formats across partner organisations
• limited analytical capacity in local areas and insufficient detail provided by high-level analyses for local partners
• insufficient funding and resources hindering data sharing efforts
• resistance to data sharing within key partners and difficulty in securing senior buy-in
• absence of a single system for data sharing across partners and challenges in mapping data within existing systems

4.3 VRU contribution to data sharing and analysis

It was evident the contribution to data sharing and analysis locally was significant in most VRUs. It is unlikely the levels of progress would have been made in the absence of the VRU. In other VRUs, there was still a contribution evident, but this was more viewed as speeding up or enhancing work that was happening already, or would have happened eventually.

This assessment of VRUs’ contributions to data sharing and analysis were broadly mirrored in responses to the Core Member Survey. When asked how they would best describe the VRUs’ contribution to data sharing and analysis locally over the past year, a third (33%, n = 53) said it was a lot more efficient and effective, and 44% (n = 47) said it was somewhat more efficient and effective. The significant minority that thought the approach was no different were spread across multiple VRUs and were typically LA or VCSE representatives.

Figure 4.1: Core Member Survey – perceptions of VRU contribution to data sharing and analysis

5. Community and young person engagement

This chapter reviews the evidence on how VRUs were engaging young people and communities beyond the direct work of the VRU-funded interventions (covered in Chapter 6). Youth and community engagement is intended to contribute to all aspects of VRUs’ response, not only commissioned or direct delivery. Aligned to the programme-level ToC, the chapter considers progress towards effective action against the following (connected/sequential) stages:

Stage 1

• a. VRUs meaningfully engage young people and communities
• b. VRU activities are co-developed with communities and young people

Stage 2

• a. VRU activities contribute to improved understanding of the perceptions and needs
• b. VRUs’ work is more closely aligned with the needs of young people and communities

Stage 3

• a. VRU activity informed by young people and communities is more effective and better able to meet needs / achieve desired outcomes

The chapter follows these stages to present the findings from interviews with VRU staff, stakeholders and frontline professionals, community representatives and young people who were involved with VRU-funded interventions. It also draws on the ethnography that was conducted at 3 sites, with a day spent observing how an intervention operated, its links with the area and its planned and informal engagement with young people, followed by one day spent in the company of an individual young person (15 to 18 years old) who used the intervention – accompanying them as they used the intervention and engaged in other activities. The ethnographers spent time on site observing planned activities or following activities as they developed, asking specific questions of staff and young people where suitable to inform their understanding and interpretation. In addition, the ethnographers held several shorter semi-structured interviews with staff and young people to obtain more detailed perspectives on the projects and interventions. When the ethnographers accompanied the young person for a day, they were again following planned or ad hoc activities led by the young person, and asking questions to understand the young person’s motivation, experience and perspective on these activities and day-to-day life. They conducted all engagement and data gathering with informed consent, including where they used filming and photography.

In total, the evaluation engaged with over 50 young people across 20 interventions at 15 sites. Of these, 24 were engaged in the ethnography and 26 were interviewed for the core elements of the evaluation. The evaluation did not seek to speak with young people involved in VRU youth panels or Boards. Although they may have a closer view of VRUs’ strategic role, these individuals are not typically users of the VRU-funded interventions across areas. The numbers of young people engaged was lower than intended, due to a combination of concerns from staff that the national evaluation replicated or added burden to young people’s engagement in local VRU evaluations and delays due to sequencing after VRU staff, stakeholders, frontline professionals and community representatives.

The ethnographers interviewed VRU staff, stakeholders and frontline professionals in all areas and with community representatives in all but 3, including those where the VRUs were just established and one with less community capacity for engagement in the evaluation. They also included intervention staff in the ethnographic case studies.

5.1 Progress towards community and youth engagement

The ToC was amended in Year 4 as previous rounds of evaluation had shown that young people and communities had limited awareness of or (therefore) positive perceptions of the VRUs and these were not considered to be fundamental to VRUs’ success. Instead, the key activities for VRUs in relation to communities and young people were understood to be meaningful engagement and co-development.

Progress towards effective engagement of young people and communities varied greatly across the VRUs during Year 4. More than half of VRUs reported good progress, and, importantly, improved progress based on previous years. However, this means that several VRUs were behind their intended progress on engagement. In some areas, there were more substantial challenges with the levels and range of engagement, including patchy coverage across the geographic area of the VRU or with specific communities and groups. In these areas, progress could feel very slow and prove frustrating for those closest to engagement efforts, both within and outside a VRU.

In all areas where young people were interviewed, they had less awareness of the VRU itself and its engagement efforts – and limitations – than the community representatives had. Even where young people had engaged with VRU-funded provision or VRU-facilitated events, they were not aware these had anything to do with VRUs. Some projects avoided referencing ‘violence’ and so left the VRU reference off project and programme materials, but overall young people were unaware these activities connected to a wider programme of work or strategic effort. Engagement activities included some co-production elements, although these were less extensive across VRUs than other forms of engagement. Co-production includes co-development and co-design, and in the context of the VRUs, was taken to include community and youth representation on VRU panels and Boards or involvement in commissioning processes. Engagement activities included some co-production elements, although these were less extensive across VRUs than other forms of engagement.

5.2 VRU approaches to community and youth engagement

VRUs take a range of approaches to engaging the wider community they cover, specific geographic and other communities on which their interventions are focused and the young people with whom they aim to achieve impact.

5.2.1 Forms of engagment

Across the VRUs there were numerous examples of forums, events, activities and certain job roles (for example, school liaison officers) which facilitate young people and communities’ opportunities to engage with the work of the VRUs. Some of these predate the VRUs but had become increasingly aligned, whereas others developed because of VRU activities.

In addition, some VRUs made further efforts in Year 4 to increase their more structured engagement with local communities and young people; for example, one VRU had recruited a new permanent Youth Lead, established a new Citizens Group and was establishing a more formal process for young people and members of the community to be part of its decision-making.

Separately, the VRU-funded interventions had different ways of engaging both young people and communities. One intervention provided an example of how they engage both young people and communities as part of delivery. The intervention engages with schools prior to the intervention starting to explain what the project is about, what type of mentoring would be received and an introduction of the services for young people. They ensure that the young person is aware of the potential of the referral. The interventions seek to ensure that the referral process is a safe environment for both for young people and the community by conducting a safe and secure risk assessment around their interventions, and try to get a holistic understanding of the young person and their family.

Facilitators and barriers

In all areas, engagement appears closely linked with multi-agency partnership approaches, especially in relation to means of engagement and methods of communication. Where multi-agency working is showing good progress, there is often good progress with the engagement of young people and communities.

Good progress appears to have been made in VRU areas with a wide range of engagement routes, from the more formal citizens panel-type approach, through to fun events held within the community and engaging with people in the youth justice system who would otherwise be reluctant to meet in a forum. Some VRUs reported that the need to offer this diverse range resulted from feedback from young people. Additionally, better progress has been reported in VRUs that have learnt from and built on engagement activities and strategies used in previous years. This underlines the value of VRUs undertaking reflective activities, formal monitoring and evaluation and using the findings of these to apply to evolving strategies and operational planning.

Conversely, in VRUs that have made slower progress in developing engagement and improving perception, there is evidence of limited communication and some indication of trust becoming frayed, including across agencies that link strategic activities of the VRUs to the young people and communities they serve. In some areas, community representatives referenced the patchy nature of VRU engagement efforts as undermining confidence in their role and relevance, noting that VRUs were less closely linked – if linked at all – with communities, areas or groups who were among the most affected by violent crime. In these discussions, those VRUs were seen as engaging most readily with groups or areas with which they were familiar or found ‘easiest’ to access. There was a view among professional and practitioner stakeholders in the areas that this could also be to do with a lack of granular data on groups showing particular need or risk, or limited use of data that could be accessed or gathered to show this.

Other barriers relating specifically to young people, noted by those involved in delivery as well as by young people, included the timing of engagement activities, age limits or VRU engagement activities being held in one area that some young people from other areas will not visit. VRUs’ association with police was also cited as being a barrier to young people’s involvement in VRU engagement activities. Whilst young people may not have been aware of VRUs as an entity, their activities being associated with police could be an impediment to engagement for some. Some interventions were specifically linked with police (see Chapter 6), but where they were not, then distrust of police could limit engagement. This could be exacerbated if a VRU was more closely aligned with Police, Fire and Crime Commissioners (PCC) (for example, by being based in their office).

5.2.2 Forms of co-production

Examples of effective engagement approaches that facilitated co-production, even if not directly doing it, focused on empowering young people and community members by listening to the strengths, challenges and priorities of the community. Some VRUs found surveys, listening projects, conferences and youth consultation activities to be effective means of including young people in the process of feeding into the shaping of priorities and spending for the 2023 to 2024 financial year. Other effective engagement activities, such as social media campaigns, were co-designed with young people to ensure they resonate with the target audience.

There is evidence of a range of forums where young people and communities can contribute to interventions or feed into local strategies. Some of the clearest examples of young people being engaged in the VRU work are where there is freedom to work with frontline providers to choose activities, such as sports or arts activities. Some providers have developed increased flexibility in their offer to allow for greater choice. This has seen some benefits, including better engagement and more regular attendance at sessions.

There was evidence that co-production engagement events have proven effective, where undertaken. For example, in one area a VRU-sponsored Hope Hack event^{[footnote 8]} was viewed as a positive means of engaging young people. This enabled the VRU to bring approximately 90 young people aged 14 to 25 together to speak about the solutions to ‘how we can create a society free of violence and a society that is fair for all’.

Co-design has proven increasingly popular in VRU-funded interventions. Young people are given opportunities to shape their own activities and interventions, allowing for greater ownership and increased responsibility in ensuring positive outcomes are achieved. There are many examples of this approach, one being a behaviour and consequence workshop, where the staff team designed the workshops but young people could design the rewards (for example, young people recently chose to attend a Premier League football game). The intervention is more positive, constructive and well attended when young people have a voice and can choose activities, which not only helps to design the programmes but helps to feed back on the challenges that they are facing.

“[In this project] you’ve got some, what other professionals would say are, ‘really disengaged young people’ that will engage very well with us because it’s theirs. They’ve designed it, that is what they want, it’s what their peers want…and it gives them that ownership that then forces them to turn up and drag other people along that wouldn’t necessarily access provision.” (Frontline professional)

Stakeholders viewed Community Leadership programmes positively, with one VRU training its third cohort. Programme members receive free training to develop skills they can apply in their existing roles to work collaboratively with the VRU and each other as community peers to build stronger and safer communities. In one VRU with an established programme, the cohorts of trainees can join the Community Leaders Network, with 2 places on the Strategic Partnership Board to provide a community perspective at the governance level.

Facilitators and barriers

The shift to the 3-year funding model was, in some cases, seen as positive for engagement. The stability of longer-term funding encouraged increased strategic focus on building local community networks and embedding support for young people and communities, with VRUs and wider partnerships playing a guiding role in local activities.

Where there has been good engagement with, and perceptions of, the VRU there has been a diverse range of activities that have included young people in their design and delivery. The greatest positive perceptions have been where frontline providers have worked with young people and communities to hold themed events focusing on mental health or with a cultural or sporting theme. From the small set of qualitative interviews conducted with young people, it was evident that even where there were positive perceptions, they did not understand the activities as being related to VRUs. Indeed, the reference to VRU was unfamiliar to most of the young people interviewed.

There was some recognition of the need to build trust and one effective way of achieving this was a ‘You Said We Did’ initiative. Feeding back to young people and the wider community on actions they or partners have taken to address issues raised in previous listening projects demonstrated they had been heard. Over time, this can establish positive reinforcement and therefore increase trust in the engagement process.

Barriers or limitations on the co-production aspect of VRUs’ work included recognition of the limited reach that some had across the community or with different sets of young people. Stakeholders and community representatives observed that facilitating multiple opportunities to provide feedback to the VRU has limited value if the main issue is outreach within the community. The challenge of getting beyond the ‘usual’ or most readily engaged participants was also not always recognised, with close associates and even VRU staff and community representatives observing that some VRUs still over-estimated the extent to which their engagement reflected the key population.

Frontline organisations identified specific challenges around engaging young people in co-production. At a basic level, this included young people not attending pre-arranged meetings or turning up too late to participate fully. Another barrier identified around co-production was sustaining young people’s interest and engagement over a longer process, which co-production would often involve, not least in finding out whether a co-designed project had been successful in a funding bid. There was also recognition that co-production may be slowed as the trauma that some young people have experienced or their complicated day-to-day lives may limit the opportunity for focused conversations with them about this aspect of the work.

5.2.3 Improved understanding of and response to need

The expectation laid out in the ToC is that the activities of engagement and co-production will lead in the near term to improved understanding of the perceptions and needs of young people and communities, and to VRUs’ work being better aligned to the needs of young people and communities.

There was evidence from the evaluation that the increased levels, breadth and depth of engagement, and the use of some additional co-production approaches across more aspects of VRU strategic planning and commissioning, had partly contributed to an improved understanding of and response to young people’s and communities’ needs. However, the evidence on this was patchy both at the national level and within VRU areas in relation to different communities, groups and areas. As noted earlier in this chapter, VRUs varied in how effectively they used data to interrogate and direct their engagement efforts, and some were perceived as being particularly distant from many of the communities they sought to support and/or making less determined efforts to reach young people who were furthest from mainstream activities.

Interviews with stakeholders found a mixed awareness of the VRUs’ own understanding of how well they are communicating and engaging with young people and communities affected by violence. On the whole, VRUs reported their levels of engagement to be better than local communities perceived. However, VRUs were sometimes unsure if communities knew what VRUs were doing and whether the outcomes would reflect community need. Overall, there was a clear understanding of the need to do this better and more systematically. Across areas, there was a clear objective among VRUs to do even more to engage young people and communities and include their feedback in the design and delivery of initiatives. In particular, there was wide recognition across the VRUs of the need to have the ‘voice’ of young people and communities affected by violence as recognisable in not only strategies and plans but also in the operational responses.

Catalysts

The research with community representatives and young people who used the interventions indicated that frontline delivery organisations can be key catalysts for converting engagement to insight for VRUs. Those that are both closely embedded and highly recognisable in their communities are seen to be effective, not only in the delivery of activities and interventions, but also as a conduit for young people’s and community member’s views. Recognised individuals with a local profile, such as sports coaches or youth workers, can also play a catalytic role in communities and with young people, relaying views between VRU and participants/communities and building rapport and trust. Over time, their role might contribute to raising the profile of VRUs, but this was considered less important than providing a key conduit or reference for understanding and responding to local needs. The 3-year funding was seen as helping to provide certainty and therefore retain/confirm staff in post and protect their local networks.

5.2.4 VRUs role is more effective and leads to better outcomes

The longer-term outcome set out in the ToC is that VRUs’ activities will be more effective as a result of engagement and co-development with young people and communities, which in turn would contribute to VRUs’ related impacts:

• communities feel safer (in terms of perceptions of community safety)
• life outcomes are improved for young people supported by VRUs (in terms of improved educational engagement/attainment, improved employment prospects and improved mental and physical health)

Perceptions of community safety

The evaluation did not provide evidence of whether perceptions of community safety had shifted due to the VRUs’ engagement and co-development activities specifically. Where there was a stronger track record or recent comprehensive efforts to engage community groups and representatives, especially in commissioning decisions, there was a more positive regard for the work of the VRUs and a greater expectation of what they might achieve. However, even in these settings, there was recognition that this perspective was not representative of communities as a whole or across the VRU area, and concern that some might benefit less extensively from the VRUs’ actions.

Young people’s life outcomes

The evaluation provided evidence that young people involved with VRU-funded interventions experience a range of outcomes related to the longer-term educational, employment and health aims of the VRUs. The young people who took part in the core fieldwork through interviews and self-completion questionnaires put particular emphasis on fundamental socio-emotional outcomes such as confidence/self-assurance, social skills and anger management. Chapter 6 discusses these as they relate to the work of the VRUs’ interventions, but this quote is a useful example:

“This has been the best place I’ve could have ever been, in that transition period where you go from a teenager to an adult…I think if there was anything I wanted to get into, these people here could help, like, start off my career.” (Young person)

5.3 VRU contribution to community and youth engagement

As outlined above, the evaluation found some, albeit patchy, evidence that youth and community engagement have informed VRU responses. Engagement with young people has developed over the past year, reaching a greater number of more varied individuals. However, there is room for progress on consistent local feedback and engaging still further, moving from engagement to co-development/co-design. The fieldwork found little evidence that young people had opportunities to engage in the strategic and operational activities of the VRUs, although they were involved in directing interventions’ delivery.

Engagement with communities was found to be more patchy still, with many doing well but some VRUs seen as connecting only to a limited sub-set of their population or areas of need. There was a consideration that the requirement for VRUs to be data focused and concentrate on formal evidence may prove unintentionally exclusionary of forms of less formal evidence and feedback provided through youth and community engagement. This in turn could add to disconnect between VRUs and those they are supposed to be engaged and serving most directly.

Stakeholders saw VRU engagement as enabling young people to hear from professionals about what is happening in their communities but recognised a need to bring young people into shaping discussions. There was evidence that VRUs were looking to increase opportunities for communities to be more involved in co-production activities and to embed their views in local activities responding to SV. One example was a VRU community engagement event on the SV Duty, discussing with community groups the impact of the SV Duty and what it meant for them engaging with CSPs and LAs. According to a community representative it was well received.

Many VRUs take a proactive lead in local evaluation of commissioned interventions and wider activities, and these are sometimes informed by co-production with the community and young people. However, there is limited evidence to suggest community and young people’s involvement heavily influenced VRU interventions. In most cases, VRUs feel the funding and top-down strategic priorities shape their interventions. As stated above, most VRUs consider further involving young people and communities in their work is a priority.

6. Interventions

This chapter sets out the programme-level findings on the VRUs’ commissioning and delivery of interventions. It assesses how effectively the VRUs were conducting their commissioning; what contribution they made overall to both the commissioning process and the delivery; and which aspects of their varied approaches proved most (and less) effective. It begins by setting out what was commissioned and the expenditure on these interventions.

6.1 Progress on intervention commissioning and delivery

A total of 327 interventions delivered support to young people during the 2022 to 2023 financial year. The number of interventions in each VRU ranged between 4 and 61, and this did not appear to correlate with the size of a VRU’s grant. Based on analysis of quarterly monitoring returns from each VRU, key points to note on the interventions delivered include:

• as detailed in Table 6.1, interventions aimed at all young people (that is, universal) were most common, closely followed by those aimed at young people at a potentially high risk of becoming involved in violence and those who were known as at risk to services
• in terms of delivery model (also Table 6.1), early intervention and prevention (selected by VRUs based on options provided by the Home Office) were the most common. This aligns with the preventative nature of VRUs
• interventions covered a range of approaches, with the most common including mentoring (n = 45), social skills training (n = 43) and sports programmes (n = 32). It should be noted that a large number of interventions (n = 95) recorded various ‘other’ as their approach, rather than those pre-specified in the quarterly monitoring returns
• over a third of intervention (38%, n = 123) approaches were those considered as high impact by YEF^{[footnote 9]}
• as detailed in Table 6.2, it is estimated that VRU interventions supported 271,783 young people. As to be expected, universal interventions reached the most young people owing to the greater number of interventions and, by their nature, these typically reaching more young people
• also to be expected and detailed in Table 6.2, interventions aimed at higher-risk groups were, on average, more expensive overall and per young person supported. Aligned to the YEF toolkit, interventions aimed at these higher-risk groups – for example, focused deterrence, cognitive behavioural therapy (CBT), and hospital navigators and other ‘reachable’ moment interventions – typically have higher costs^{[footnote 10]}

Table 6.1: Number of interventions delivered by target group and delivery model, all VRUs, 2022 to 2023

Target groups	Prevention	Early intervention	Therapeutic	Desistance	Not specified	Total
Level 4: Involved in crime/violence	3	4	8	32		47
Level 3: Known risk/Known to services	10	27	22	17		76
Level 2: Potentially high risk	15	49	19			83
Level 1: Universal	51	24	19	4	4	102
Not specified		1	2	1	15	19
Total	79	105	70	54	19	327

Source: Quarter 4 monitoring returns

Notes:

1. Excludes one VRU and interventions not recording supporting any young people, such as those covering professional development.

Table 6.2: Number of projects delivered, total spend to date (April 2023) and number of individuals supported, by target group, year ending March 2023

Target group	No. of projects	No. of young people supported	Mean cost per young person supported	Mean cost per young person supported
Involved in violence/crime	44	20,039	£1,556	£737
Known risk	68	59,141	£1,173	£682
Potentially high risk	67	49,280	£1,024	£323
Universal	94	105,497	£337	£116
Not specified	15	37,826	£748	£398
Total	288	271,783

Source: Quarter 4 monitoring returns

Notes:

1. Excludes one VRU, and interventions not recording supporting any young people or spending any funding, such as those covering professional development. The latter explains the difference in the total number of interventions between Table 6.1 and Table 6.2.

6.2 VRU approaches to commissioning and delivery

VRUs’ strategic oversight and local engagement are intended to inform their commissioning, so we expect the bespoke set of interventions delivered in any year to be:

• strategically commissioned and evidence-based
• appropriately targeted and effective in supporting those most in need
• lead to the intended outcomes – both for young people and for the wider community

This section considers the extent to which progress towards these objectives was evidenced in Year 4.

6.2.1 Strategically commissioned and evidence-based

Strategic commisioning

In this fourth year of operation, and with the move to multi-year funding, there was evidence that VRUs were more established in strategic commissioning. There was still variation in the extent to which VRUs engaged partners in commissioning decisions, and a consequent lack of clarity for some over the process. However, as a set, the VRUs demonstrated they were moving to consider commissioning from a strategic perspective, with sight of the overall landscape of provision and need, rather than primarily as a booster funding stream. One professional in a strategic role described the VRU as having “instilled the thinking of systems change rather than systems funding”.

Hallmarks of good progress included the close engagement with partners or other agencies to align commissioning decisions – or at least to check against duplication of interventions – and the use of multiple sources of evidence to provide insight and guide decision-making. Section 6.3 discusses how VRUs’ contribution to commissioning and delivery can be mapped against their approach to partner, multi-agency, community and youth engagement to understand the VRUs’ role in influencing the range and structure of local interventions. It was evident across the sites, however, that they saw their commissioning role as strategic:

“We set out a criteria, so a lot of it is based on or aligned to a strategic priority. So, one of the 5 themes plus the trauma-informed work, multi-agency work, whole-system work and it being within [multiple] communities, but it has to align to one of those strategic outcomes.” (Operational stakeholder)

The 3-year funding was seen as a catalyst for this progress, enabling VRUs as a set to commission and award grants in a more structured and collaborative process and to take a longer-term view on impact measurement and what can be achieved across the system.

Evidence-based

The requirement for VRUs to commission based on evidence and identification of need was well-recognised, with a view among professionals that it is core to (even “ingrained in”) the VRU approach. Key evidence across all VRUs was the SNAs, local data and feedback from young people or communities. The YEF toolkit was also cited in most sites, although familiarity varied by area. The need to know what is happening on the ground, even if not evaluated, was highlighted as a key commissioning premise for VRUs. One area was placing increased emphasis on being alert to different ways that ‘violence’ can show up in young people’s lives. There was appreciation from those delivering services that the VRU had recognised this broader interpretation of violence, allowing for day-to-day observations of lived experience to be used as evidence in commissioning:

“Originally, young people who’d been admitted [to A&E] had violence but also mental health issues, injury from sports – don’t always see what else can be going on – can still be a reachable moment. Most at risk of violence or exploitation have gone with substance misuse linked to mental health… The amount of young girls with overdose, self-harm, some so socially isolated have no one else and A&E is the only place they need to go, when really they need someone… If [the VRU] had stuck so closely to violence incident, would never have reached all these people.” (Frontline professional)

Additional sources of evidence in VRUs showing greater progress included more formal or systematic use of local insights such as listening projects, closer engagement with partners / their data, reference to Response Strategies, tracing trends, a matrix to prioritise interventions, or more analytical insight from academics, pilots or evaluations.

“We [the VRU] have done a needs assessment which we’re just updating, we’ve got our Response Strategy, we’ve got the online portal where they can see police, ambulance and hospital data overlaid which really helps CSPs in terms of understanding the local issues. We’ve commissioned analysis reports from [academic partner] which help us with that.” (Operational professional)

VRUs showing the greatest progress put particular focus on evidence to inform immediate or future commissioning. Where evidence did not exist, these VRUs sought to develop it through their commissioning. It was acknowledged that demand for evidence or outcome monitoring can burden smaller organisations, and that this can be an issue for commissioning them, given the expectations for data gathering. However, there was a view that they can be highly effective on less formal measures and do meet other VRU model requirements, such as the engagement they have with communities and the extent to which they represent the communities they deliver to.

Additionally, as outlined in Chapter 5, there was expectation that VRUs should be drawing on young people’s and communities’ insights and perspectives as ‘evidence’ or at least as relevant information in determining what should be commissioned, and how commissioned interventions should be delivered. There is a question over the extent to which it is appropriate to premise VRU direction and delivery on community and youth engagement, but there is certainly a role for both to be drawn into formative and reflective processes of VRU strategy. The evaluation found that VRUs are progressing on this, but that efforts are patchy at both a national and local level. At the intervention level, however, young people’s perspectives were more often actively sought and integrated into activities. Interviewed participants reflected on staff encouraging them to give their feedback on activities. They valued the request for feedback for making them feel listened to, and the young people appreciated their feedback being taken on board, as they saw this translate into changes to activities:

“It has been quite nice to actually have our voices and opinions be heard, and actions taken.” (Young person)

Use of YEF toolkit

The YEF toolkit is expected to inform commissioning^{[footnote 11]}. The toolkit is an overview of youth-focused interventions and approaches to preventing youth violence, with summary reviews of the evidence base and strength for each. The YEF describes it as “a free online resource to help you put evidence of what works to prevent serious violence into action”^{[footnote 12]}. Most VRU sites mentioned the toolkit, but for varied uses. Some used it for broad reference and signposted partners to it. Others referred to it for set technical activities such as intervention monitoring or impact evaluation. Others used it to inform commissioning in general, not only for the high-impact interventions which it identifies, although there were differing views on how suitable it was for this. These included interventions for which the evidence of impact on SV was deemed to be ‘inconclusive’ in the YEF toolkit (the toolkit did not include interventions for which there was evidence of no impact). It was also seen as a Home Office commissioning-requirement, a broader guide to commissioning across the range of interventions or a validation of local choices (one stakeholder noted that over four-fifths of local interventions were in the toolkit). One site used it to structure their co-design of interventions ‘from the start’.

However, there were also comments across the same sites on its limitations and prescriptive approach. Whilst considered a useful reference for commissioning, it could be seen as a relatively ‘crude’ tool beyond this. It was also said to be too focused on randomised control trials (RCTs) to aid local/community discussion of commissioning or to inform the commissioning of smaller providers. It was acknowledged, as set out in the Delivery Plan template, that VRUs could commission interventions that are not in the toolkit, as long as they then evaluate the intervention (although the delivery plans showed that not all were intended to be evaluated). However, there was also a view that the way VRU options are framed by reference to YEF standards can hamper community involvement in decision-making, with an explanation that there are only a limited set of interventions that can be funded and these may not be what areas want.

Commissioning of high-impact interventions

High-impact interventions are those the YEF toolkit identifies as estimated to have high impact on violent crime, although the VRU Delivery Plan template also provided for VRUs to estimate the impact for interventions which were not in the YEF toolkit, and this did include those estimated locally to be high impact^{[footnote 13]}. Examples include focused deterrence projects, hospital/A&E navigators programmes, CBT and early intervention efforts such as certain school-, sport- and community-based provisions. In Year 4, following discussion with VRUs and the YEF, the grant agreement included a new requirement for at least 20% of each VRU’s intervention budget to be spent on interventions the YEF consider high impact.

Although identified as priorities for this latest round of funding, and highlighted within delivery plans, professional stakeholders across the sites did not readily identify interventions as high impact in interviews. Those who did said that most interventions followed on from earlier years, with some being commissioned additionally in areas. It was evident from delivery plans that (re)commissioning of high-impact activities was considered to be relatively straightforward as these interventions were considered to have proved successful in previous years. Typically, these were YEF-defined/aligned high-impact interventions, but areas did also reference their own experience of some interventions as evidence of high impact in the plans. This meant that the YEF toolkit and funding requirement may have reinforced but not led decisions around many of the high-impact interventions. Next year’s evaluation could consider local understanding of ‘high impact’ in terms of the YEF toolkit or other specification.

Contributing to the evidence base

The VRUs are intended both to draw on and contribute to the evidence base, by evaluating their interventions and sharing findings back locally and beyond, as well as drawing on these to direct/redirect their attention, funding and activity. Beyond the local evaluations conducted as a condition of the central funding, there are programmes of local research, piloting and innovation which could potentially provide evidence beyond the area. In this fourth year, there was substantial progress made collectively by VRUs in contributing to the evidence base.

Despite overall progress, the extent to which VRUs had achieved systematic monitoring and evaluation still varied. One cluster had made limited progress, with an indication of ‘positive intentions’ to contribute to the evidence base or lack of clarity over the extent to which analyses would be shared back through the VRU to its wider stakeholders. Among these VRUs were those that predominantly drew on interventions’ own monitoring, which was less suitable for wider use. A second cluster of VRUs had made more progress and had established performance and outcomes frameworks, or were providing monitoring and evaluation input and advice for other organisations or for their own system-wide view of need.

A third cluster, the largest set, used their understanding of the evidence to contribute to it further. Among this set were VRUs that had commissioned feasibility studies; developed, run and monitored pilots; initiated programmes of research; and expanded their monitoring to capture more than the Home Office required. This set also included those with larger programmes of evaluation and those working to embed monitoring and evaluation systematically across their interventions. One VRU had developed an Intervention Monitoring Dashboard and embedded it across all interventions, and had shared dashboards that partners drew on for their own monitoring and evaluation.

The fact that core funding was now assured for 3 years was seen as enabling larger, longer-term evaluations, rather than jumping from pilot to pilot. However, there was also a shared observation that other funding sources can be crucial in enabling VRUs to add to the evidence, as it can be less restrictive than core funding might be. Examples include a pilot project which had gained ‘national traction’ and shown partners how it can be replicated.

“It’s not a great deal of investment, but just sharing how that can be done through our small commissioning projects, actually recognising and sharing the results of that [has helped build the evidence].” (Operational professional)

6.2.2 Appropriately targeted and effective in engaging the most in need

VRUs’ commissioned interventions are intended to be:

• targeted at individuals and areas identified as most in need
• effectively engaging the priority groups, including through high-impact interventions

Targeted at individuals and areas identified as most in need

VRUs do draw on data, local information/feedback and wider evidence to identify places and groups of people as most in need, but it is less clear how consistently they can target their interventions at those people and places.

There was broad confidence that areas most in need were being identified annually, through the combined use of administrative data and local knowledge. However, there were observations in some sites that areas perceived as higher need locally are not prioritised under the current system and concern that rural areas can be overlooked. Those making these observations included community representatives and non-VRU professionals, with some recognition from core staff that there were challenges in ensuring that high-needs areas are fully recognised, if they are areas where individuals reside but not where the SV problems show up. Examples included where a night-time economy area or central urban area is identified as a hotspot for SV but the patterns of residence for those involved is not routinely assessed or responded to in the same way.

It was noted that VRUs focusing on pan-site interventions could miss some vulnerable communities (and groups – see below) that were seen as priorities locally by frontline professionals and community members. In other sites, the data-driven approach was seen as having obscured areas until later: “lot of investment gone into a couple of areas and then last minute for the others”, as one professional noted. There was more active concern in some sites that the VRU was failing to engage appropriately with areas and groups of highest need, but overall there was some progress in this respect. Box 6.1 has an example of successful use of data in identifying target groups.

Box 6.1: Case study example of data-informed focus on highest need

Through its Year 4 data analysis, the VRU was able to identify 10 to 14-year-olds as the highest risk of SV, which highlighted the importance of their schools network and education leads work in connecting education data and partners into the VRU. In addition, it began showing evidence of impact against the intended outcomes (short-medium term) of its re-commissioned interventions and working with delivery partners to further refine the targeting and engagement capacity of interventions.

More VRUs were making progress in tracking geographical changes in need. In some sites, multiple data systems were combined to help in identifying local trends and hotspots where issues ‘flare up’, such as in night-time economy areas or locations where problems arose after school. The ability to identify and respond to hotspots was important for policing and useful for interventions with flexibility in where they operated, such as detached youth work. The data was less suitable or responsive in identifying where those associated with shifting issues might live or whether these locations would constitute areas of high need if identified.

There was a recognition that VRUs could identify key priority groups for interventions through combining data, local knowledge and broader understanding of trends and issues. In some sites, this information was refreshed and reviewed through the use of dashboards or other live data sources and ongoing or occasional feedback processes. In others, the picture was determined more rigidly from the SNA, although the partnership-working and partner referrals provided further opportunities to identify those most in need. In some areas, it was considered that the definition of community could more actively include faith-based groups, and one community organisation in a site thought its VRU should look to work with inter-faith groups. There were also recognitions of ways in which VRUs were responding proactively to prioritise communities that may be out of the wider view, including newly established or growing communities:

“So what the VRU, I think, has done, especially with the funding they’ve given to us, is let the group that we’re working with from that community be part of something bigger, something established [in the area]. So it allows others to see that this community aren’t just here to work and that sort of thing, they’re getting involved.” (Community representative)

There were questions over how comprehensively VRUs prioritised all of those most in need, with a broad agreement that the core groups were identified but various gaps remained. There is a strong evidence base on key risk factors that are associated with SV, much of which has been available from the outset of the VRU programme (see for example, Smith & Wynne-McHardy, 2019). Nevertheless, the groups that were overlooked varied by area – discussed below – but indicated that VRUs are seen as simultaneously missing both those earlier in the trajectory and later.

There was a widespread view that a public health approach to SV requires still earlier identification and engagement to ensure it is preventative. One VRU had identified children aged 10 to 14 as being most at risk of SV, but stakeholders in other sites doubted interventions were targeted early enough. Children with parents/siblings in prison or known to be involved in crime or violence were identified as a particular gap, but overall the concern was to focus on children earlier. Running alongside this concern was the observation that VRUs were not sufficiently prioritising young adults and young people who are already entrenched in offending and violence, including younger prisoners and young people leaving custody. This was a widespread concern across the VRU sites, with challenges described in different forms in different areas but a recognition that those who are closest to offending and SV can be differentiated into sub-groups and that areas had gaps in the groups that they prioritised.

Effectively engaging the priority groups, including through high-impact interventions

The challenge for many VRUs was moving from identifying priority groups to making sure that the interventions were appropriately ‘targeted’ on these groups’ needs or interests and could therefore effectively and consistently engage them. Opinions were mixed on this. There was a strong narrative questioning if VRUs could interpret the evidence appropriately given what was perceived to be, in some sites, their narrow frame of reference. Whilst the data may show what the issues are, and where these issues are located (geographically, by community or group), it was explained that in some instances a VRU’s ability to determine what is the optimal engagement approach could be constrained by not having a deeper understanding or familiarity with the end users:

“The delivery, the learning, the lack of understanding of what the community really needs and if I’m brutally honest, in regards to the VRU that I’m aware of, the absolute lack of diversity there… For me, there are too many people in the roles at that organisation that have absolutely no idea about the community. This is why I say they then fail when it comes to commissioning the right groups.” (Strategic professional)

It was also noted that monitoring data could not necessarily show the engagement levels or profile accurately. Levels could be under-reported where young people refused to consent to monitoring (in case information was shared with police) but remained engaged. In other cases, the profile of young people who engaged might not be what had originally been prioritised. An example would be young people involved in anti-social rather than criminal behaviour engaged in projects aimed at those entrenched in SV. A frontline professional noted that who is engaged can be influenced by settings. Their universal primary prevention work tried to target set schools “but ultimately you have to have go into schools that accept you”, although they noted that “anyone can be at risk”.

However, despite the reservations or limitations, there was clear evidence that engagement was proving effective. High attendance levels or waiting lists showed high engagement across many interventions. Examples given included sustained attendance across all 20 sessions of trauma counselling in one site. The intervention-level monitoring in multiple sites was considered useful for tracking engagement, including for intervention staff:

“It’s making problem solving so much more effective and it’s so much more agile and fluid, but it also means that we can say, ‘Look how incredibly you are doing, you’ve got literally 82% engagement rate with young people who have been referred’. That’s incredible.” (Operational professional)

The fieldwork with young people explored how and why they had engaged with the interventions. Referral routes were mixed, and included formal referrals from school, prison or key-workers, and informal or self referrals and just ‘dropping in’. A key theme relating to their initial expectations of the initiatives centred on expecting that the intervention could improve their social skills and help them make connections and friends in their community. Another theme comprised them feeling that the initiative could support them with future employment opportunities, such as providing them with new skills to add to their CV. For initiatives that were sport-based, young people said they had been looking forward to the physical activity, and for it to help them feel more fit, active and confident.

There was agreement that interventions had surpassed their expectations, with young interviewees frequently reflecting positively on how they had felt after attending the first session. As they attended more sessions, the young people interviewed commented on enjoying feeling increasingly more involved, taking part in more activities, and committing and persevering (especially in sports-based interventions where the activities might feel particularly demanding). Aspects of interventions which were particularly appreciated were having freedom and choice in what activities they took part in as part of the interventions – as highlighted in the review of co-design in Chapter 5. Having smaller group sizes, 1-2-1 support or where learnings could be applied in the sessions or soon after were all also seen as beneficial, making them feel more comfortable and get more from an intervention.

“I was like, ‘I don’t know if this is my thing.’ Then when I first went I was like, ‘This is my thing.’…I loved it…, it was really scary, but then everyone was, like, welcoming me in and stuff.” (Young person)

Section 6.3 outlines the most effective approaches, but this year’s focus on high-impact interventions proved more difficult to trace. It was not a widely recognised term for the VRU-commissioned work, and some stakeholders explained that interventions were not discussed as being high impact. Some sites were unable to identify young people as specifically linked to high-impact interventions, although those with more comprehensive intervention-level monitoring could distinguish this. In other sites, the term was specifically associated with tertiary or ‘reactive’ interventions. This did not reflect the range of ways in which high-impact interventions were being used across several areas but did indicate how they were perceived when seen.

One area which had commissioned several more preventative high-impact interventions – as in the YEF toolkit – found the concept could still be opaque. A senior stakeholder explained it could be a challenge to translate a high-impact model of intervention into the particular circumstances and structures of that PFA. Other stakeholders across the site shared this opinion, observing that it was not always suitable to generalise from what worked in one area to transfer to another. This does raise issues about the fidelity of intervention delivery to the underlying model as developed, but there was no specific discussion of this other than to recognise, as noted above, the challenges of trying to deliver ‘as developed’ where the context does not lend itself readily to this. The evidence on how high-impact interventions did deliver is discussed alongside other approaches in Section 6.3.

6.2.3 Lead to intended outcomes for young people and communities

VRUs’ commissioned interventions are intended to:

• realise immediate and short-term outcomes for young people and communities
• achieve the ultimate outcomes of reduced violence and improved community safety

Immediate and short-term outcomes for young people and communities

Immediate and short-term outcomes vary between sites and interventions. Taken as a set, however, there was evidence of VRU-commissioned interventions delivering on their intended early outcomes for young people, although less evidence to date on this for communities. Much of the evidence was reported for individual cases, but the monitoring and evaluation systems used in some sites provided data that reinforced the qualitative insights.

There was a small cluster of VRUs where professionals considered it too early to comment, conscious that monitoring systems were still being set up or outcomes being specified. One stakeholder said they could detail the impact – by group, location and how it was achieved – “but at the moment, it’s all quite anecdotal”.

Other VRUs could point to increasing volumes of qualitative and quantitative evidence of interventions’ effects. Evidence included monitoring data, case notes, observations from professionals, and informal or structured feedback from young people, and some from parents. Not all interventions were covered and local monitoring data are not analysed for the national evaluation. However, in a majority of VRU sites, stakeholders identified a cross-section of interventions that, in their view, demonstrated immediate/short-term outcomes. The sets of immediate and short-term outcomes being achieved, for at least some people, related to several different domains:

• connection (trust in service providers, being signposted to other support services)
• offending (engagement in re-offending and/or anti-social behaviour (ASB), proximity to gang involvement, risk of criminal exploitation and engagement in risk-taking behaviours)
• education and employment (educational engagement or risk of exclusion, access to or engagement with employment, technical or other skills, achievements or awards)
• social (attitudes to violence, self-confidence, leadership and teamwork skills, feelings of safety and belonging, respect for adults, sense of resilience, self-regulation)
• wellbeing (mental health, drug use, housing, including access to support with these)

The mix of outcomes indicates VRUs are commissioning or delivering interventions that will have more durable impacts on young people’s prospects – reducing engagement with offending but also shoring up broader life skills. One operational professional noted the data show reduced offending, but staff are also seeing and hearing about these wider outcomes:

“Sustainable life skills, to be less angry, to have better relationships, to want to go to school and want to achieve. That’s what the young people are telling us.” (Operational professional)

The young people interviewed for the national evaluation were consistently clear about the meaningful ways in which their involvement in the various interventions – and they were varied – had brought real change to their sense of wellbeing and future prospects. Summarised below are the key impacts identified by young people:

• interventions helping them feel confident in future career prospects, through the networks they had built through interventions, and by having evidence of attending courses they could add to their CV

“This has been, like, the best place I could have ever been, between that process transition period where you go from a teenager to an adult, because you get to do adult things, but you also get to lark about…I think if there was anything I wanted to get into, these people here could help, like, start off my career.”

• improving their social skills, relationships, sense of social anxiety and confidence, both now and in the future. Examples included teaching young people how to better manage conflict and de-escalate situations, and also facilitating young people making new friendship groups in a safe space, which was particularly highly valued

“The first couple of times I came here I was actually very nervous. I really struggled with, like, even here talking to people. And basically, what helped me open up, was…a crocheting activity and that helped me, like, be more open.”

“I feel really confident now compared to what I was before… I can walk out of my house knowing I’ve learnt something.”

• better managing anger and stress and improving focus and mindset. This was something young people could see interventions had done for both themselves and other attendees

“When I first came here, like, my anger was - you’d say one thing to me and I’d probably switch on you like there was no tomorrow. But since I’ve been going to [the intervention], it’s got a lot better. I know how to control it more and I’m more calm when I speak to people.”

• young people identified intervention facilitators being welcoming, motivating and supportive as key to facilitate their engagement with the initiative

“I loved the staff there, they were so motivating. I was quiet at first though because I didn’t know anyone. They were really motivating and supportive and stuff…they actually push you to your limits…when you do exercise, like, cheer you on and stuff… Makes you feel happy, it makes me actually want to do it more.”

• the sample of young people provided limited but clear evidence that the interventions had taught or helped them with something, and no evidence of these interventions failing to deliver on this where they were expected to assist. When asked, young people across the interventions agreed they would recommend the projects to others as they felt they could help others in the ways the initiatives had helped them

“Yes, I would literally recommend it to everyone…It’s changed so much with me…I see a big difference with my body…the way I act, the way you feel.”

Ethnographic case studies

The research team trialled an ethnographic case study approach to understand how VRU-funded interventions are embedded in communities, and observe how young people interact with the interventions. Ethnographic researchers spent 2 days in situ, including one day following a young person outside the intervention to see how their participation relates to their wider life. The researchers used filmed participant observation that included observing interventions and informally probing the service providers, young people, and sometimes their families and friends. The team then analysed this footage and field notes based on:

• what people said, including their values, beliefs and perceived impact of the intervention
• what people did, including observing how they interacted with each other and their level of engagement
• the context including the physical space, the location in relation to the wider community, time and external factors that may influence engagement
• how the intervention related to others operating in the community

They completed 3 case studies. This section reflects on how VRU-funded work is being implemented in 3 communities and observed impact on some young people’s lives.

In the first site, the interventions were provided through a Pupil Referral Unit (PRU)^{[footnote 14]} and a boxing academy which pupils attended regularly. In the second site, the interventions were provided through a multi-function and multi-site youth centre. In both, the staff, associates/volunteers and young people understood the key activities as being intended to support the young people over the longer term, as well as occupying, informing and supporting them in the present. In the PRU, there was a clear focus on reducing violence and ASB within the setting and beyond. In the youth centre, the concept of violence reduction was not expressed as being at the core of the work, but it did expressly focus on providing young people with opportunities for positive behaviour. In both, several key objectives were identified in discussions which align closely with the expected intentions of VRU-funded work: establishing a sense of safety; building confidence; developing life skills; and being future-oriented.

In the first site, the leadership noted this shift from a culture of aggression to a reduced sense of violence with new young people arriving at the school and ‘feeling the atmosphere’ of support and positive opportunities. Time spent on site and following one young person for a day provided evidence of how the VRU funding was used to reinforce positive engagement and provide extension activities which facilitated young people’s ongoing development of self control, sense of safety and aspiration to move away from a trajectory which might involve them further in violence.

“I feel like all the programmes that [the PRU] do are there to help us learn more and, like, be better. Like they do all this stuff to help with our behaviour. I started doing combat because I went with the school and really enjoyed it, so started doing it after school as well. I like it because of the support I get from all the staff members. They push us beyond our limit and teach us how to keep our anger in and then let it out on a punching bag later.” (Young person)

In the second site, the range of activities and the youth-led approach to initiating and tailoring what was on offer were identified as key features that aligned with the goals of the VRU funding, to provide positive engagement and co-designed/co-developed activities for young people in a community experiencing increased levels of violence. The on-site visit and time spent with a young person showed how the VRU-funded activities were valued by those the ethnographers engaged, and attracted large numbers of young people. In the words of one of the young people:

“[It] helped me discover my potential and gave me a sense of purpose. I am driven to make a difference, and with the skills and support I have gained, I have the power to do so. It’s not just an intervention; it’s a springboard for my dreams and aspirations.” (Young person)

Ultimate outcomes of reduced violence and improved community safety

Interviewees across the areas considered it too early to say what impact interventions were having on the longer-term outcomes of reducing SV and improving community safety, for young people or for communities. One stakeholder reflected a wider view that it will be “difficult to measure success until a generation’s time”.

There was also caution at what could be measured or attributed to VRUs specifically. In one site, an operational professional noted that reduction in SV was showing in the data but was still cautious about claiming this was directly due to the VRU. Professionals across the sites noted the challenge inherent in attributing final outcomes to VRUs’ efforts – given the complexity of many of the young people’s lives, the funding timeframe, that much of the investment was for preventative activity, and there were other similar/related efforts.

A cross-section of stakeholders from across different sites did question if the VRU as a model can ‘get to the root of the issues’ and address the factors that drive violence. This broader consideration is discussed in Section 6.3.

However, there was broad agreement that the logical flow from VRU-commissioned activity to short-term outcomes meant early indications of progress should contribute to the later and ultimate outcomes. There was also broad confidence in many areas that interventions were on track to contribute. Intermediate outcomes were referenced as indicating this, including reductions in re-offending and school exclusions, and increased engagement with education and interest in employment. In one area, the focus of discussion was on how more can still be done to maximise impact through more frequent and effective joined-up working.

There were also some indications of impact on ultimate outcomes, in the assessment of the VRU’s own evaluation and monitoring, although these were only reported to the evaluators and may not align with the analysis conducted at national level across the programme as a whole. As an example, core staff from one site explained they had seen a 25% reduction in homicide, 12% in all violence with injury and 12% in knife crime since baseline. They explained they thought this showed the VRU was contributing, although they caveated that they could not specify to what extent the VRU was specifically responsible. In another site, a stakeholder credited the VRU’s funding of youth provision had contributed to a reduction in ASB and another identified targeted VRU initiatives as contributing to a reduction in knife crime. The perspectives of those young people who were interviewed affirmed that they felt more confident about their personal safety and ability to hold back from being involved in violence as a result of their engagement with the interventions, supported by the ethnographic case studies.

6.3 VRU contribution to commissioning and delivery

VRUs occupy a key role in the commissioning and delivery of interventions across all areas, but the nature of their contribution varies by site, depending on factors such as their structure and linkages with local organisations.

6.3.1 Contribution to commissioning

As in the third year, a VRU’s model (centralised, localised, hybrid) could be seen as one influence on its approach to commissioning. Again, VRUs varied in how centralised or decentralised they were in commissioning, from fully centralised (one had restructured to commission centrally) to having spokes lead the commissioning, or commissioning area-wide interventions from the centre with more localised provisions being commissioned by CSPs or spokes. In addition, but operating independently of the VRU model, were 3 contributory styles ranging from more to less open or engaged. There were indications that the VRU’s contributory style may be more a function of maturity than of model. The 3 styles of contribution were:

• collaborative: engaging partners and other organisations in commissioning decisions
• facilitative: building the capacity of other organisations to collaborate over decisions
• directive: providing other organisations with ways to engage in VRU-led decisions

Directive ways of working were demonstrated where a VRU created limited opportunities for joint commissioning, or circumstances had constrained how the VRU engaged partners in the commissioning process. This happened in only a small minority of VRUs that were viewed as having made little progress on commissioning strategically. In these areas, professionals expressed expectation or hope that other organisations might be more integrated. It was not clear whether there was any association between these sites’ commissioning being both less collaborative and less strategic than others, but next year’s evaluation could consider to what extent the sites show progress in one or both dimensions, to assess if these dimensions of commissioning are linked or independent of one another.

Facilitative ways of working recognised the disparity between a VRU and other organisations in the commissioning process and sought to proactively engage them and also to build capacity across the sector, or at least to be responsive to alternative views. Strategic and frontline stakeholders described around a quarter of the established VRUs as operating in these ways. An example of a more advanced facilitative structure would be the with-and-for communities’ framework, which one VRU had implemented to drive future co-production and partner relations, which was initiated with an engagement event including partner agencies and community groups.

Collaborative ways of working were the most common across VRU commissioning. Although specific processes varied from site to site, the approach broadly involves both a wide range of partners and multiple or continuous engagement in planning and decision-making. Community representatives in several areas reflected on how community groups had been actively involved in aspects of the VRU’s commissioning processes, and in some areas, taking a more integrated co-production approach to planning the delivery.

Box 6.2 outlines an example of highly collaborative commissioning; this approach would be expected to compensate in part for the limited diversity of some VRUs, providing additional perspectives on what was needed for particular areas and local populations that staff may be less readily connected into or understand in detail. Nevertheless, we cannot underestimate the challenge of including the wider cross-section of community representation and young people’s perspectives even in a collaborative approach to commissioning, so processes of community and youth engagement would need to extend beyond commissioning to help the VRUs ensure the interventions selected are targeted, delivered and taken up by the groups who are most in need.

Box 6.2: Case study example of highly collaborative commissioning

All the partners report the VRU worked with them throughout the commissioning process, including to ensure complementary design at the outset and engagement of stakeholders to support delivery at the mobilisation stage. There was a shared view that the VRU does not commission in isolation, and was reported by operational and strategic stakeholders and reflected in the responses of frontline staff and in the operating models of interventions.

6.3.2 Contribution to delivery

The VRUs commissioned 327 interventions across all sites in 2022 to 2023, drawing on the core funding, matched or co-funding (such as from YEF) and bringing in a wide range of additional funding. As in previous years, the role of VRUs in contributing to delivery was substantial, but for the first time it could now be framed as multi-year. There was appreciation that the multi-year funding allowed VRUs to move from short-term piloting to commissioning and evaluating delivery over the longer term. The timeframe was also seen as helping provide evidence for the future:

“It’s not just that they’ve given us money for 3 years. They commissioned the evaluation to say, ‘Well, actually is this value for money? And what that then led us to do is remodel based on evidence.’ … That’s a huge piece of work and it’s going to be shaping what we do going forward.” (Operational professional)

VRUs’ networked structure and strategic oversight could make yet more of their contribution to delivery, where they were actively linked into other partners’ resourcing and aware of the gaps in their own provisions or reach. Sometimes this was about drawing on the local profile of community-based providers to add some distance from the police, with which VRUs could be closely associated, and reassure young people that the intervention could be trusted. In other cases, it was about working directly with police resources, as a VRU staff member explained:

“We have victim coordinators employed by police in each district. So, I work with those as well, to make sure there are services available and young people are more aware of those services. So, although I can’t tell them what to do, I can work with them and say, ‘Look, we’ve got gaps in service here, what can you do to try and bridge that gap?’”

6.3.3 Most effective approaches to commissioning and delivery

In this fourth year, it has become more evident that VRUs can catalyse their contribution to delivery by combining 4 key perspectives: the data insights from dashboards, partners and intervention-level or VRU-wide monitoring (to understand the issues, the need and the gaps); the independence of community-based or third-sector projects; the specialist resourcing and scale of statutory providers; and the strategic perspective of the VRU as a network. Where these are harnessed, the interventions commissioned and delivered demonstrate encouraging short-term outcomes and trajectories of positive impact in the longer term. These ingredients do not always come together, however, and so this final section outlines those individual approaches which prove most effective.

On the commissioning front, as noted throughout, there has been broad progress in the effectiveness of the VRUs’ contribution. The Core Member Survey of professionals demonstrated this, with no responses indicating that commissioning has got worse, and three-quarters of respondents noting it has improved somewhat or a lot (see Table 6.3).

Table 6.3: Responses to the Core Member Survey question “Overall, and focusing on the past year, how would you best describe the VRU’s contribution to commissioning interventions to reduce violence, locally?”

Commissioning has improved a lot	25
Commissioning has imprved somewhat	52
Not realy any different	24
Commissioning has got somewhat worse	0
Commissioning has got a lot worse	0
Not sure	6

The key elements of effective approaches to commissioning were essentially having the right information (formal data and informal insights) and networks to identify, understand and respond accurately to priority groups’ needs. As one operational stakeholder explained, the process is enhanced by:

“…linking it back to the SNA and the wider YEF toolkit as well so it’s not just blindly commissioning interventions that we think we need to commission, it’s being endorsed and spoken about at local violence board levels and consultations from partners”.

These partner consultations were enhanced where they were contextualised by meaningful community engagement and young people’s contributions. The new multi-year funding highlights the importance of taking a whole-system approach.

“I think we’re all excited by the idea [that] our violence prevention system has £[X] million worth of investment … [So] how are we gonna use that to create a more whole-systems approach? … We’re kind of at the beginning of that journey, in terms of being more sophisticated around our joint commissioning…in terms of thinking about the whole system together and what that investment looks like and where the gaps are.” (Strategic professional)

The key elements of effective approaches to delivery were more varied, but premised on the same need to ensure that what is being done engages with the right people and recognises what cannot be known from data or toolkits.

Increased outreach was recognised as contributing to the overall delivery success, beyond individual interventions. This included widened engagement with schools, where many originally did not want to get engaged with the VRU. There was a suggestion that VRUs putting “autonomy back in the hands of projects” (frontline professional) would increase young people’s engagement still further. This aligned with a clear and widely shared strategic professional view that engagement would be catalysed more effectively by using people who were familiar with locations and lived experience of the young people who were being prioritised. In addition, more rapid, extensive and/or durable engagement was reported for interventions that were co-designed or delivered with young people themselves:

“[the approach is] effective because it has young person’s views, because that’s so important for it to be successful. You need to hear from them. If you don’t get it, then it’s not going to be as effective because you haven’t got the young people’s buy-in.” (Operational professional)

In addition, a broad portfolio of interventions was considered an advantage in facilitating engagement. Examples included the locally embedded nature of some interventions – detached/street-based youth work, or community-based projects where staff live in the area – and one VRU’s life course approach which meant families could be identified and engaged at various points, enabling both the children/young people and adults to receive support. Working with a range of partners or taking referrals from multiple sources also increased the chance of engaging the right people, including partners that could overcome the potential barrier of the VRUs’ affiliation with the police. The role of community groups and intermediaries was widely recognised as pivotal here, as outlined in Chapter 5.

6.3.4 Recognising the limitations of interventions

Across the areas, professionals working at all levels were concerned not only with what VRU-commissioned work could do in the lives of young people and for their communities, but also emphasised the need to recognise what it could not do. In part, this was about ensuring that the broader strategic work of the VRU delivers, so that discrete efforts to address re-offending through sport or therapy are connected into a more supportive context where wider issues and pressures relating closely to (re)offending have also been addressed. As one professional commented:

“There’s a gap around reducing re-offending. However, it feels like these issues are bigger than the [VRU]. It’s about embedding those systems into reducing re-offending, like housing options, education, employment, skills, the CPS.” (Operational professional)

Other professionals were clear that the challenges, and the work of the VRU, should be seen as still broader yet.

“How effective as a model? I think it’s a really good model, but we have to look at the drivers of violence. A lot of the drivers of violence come from poverty and inequality. So, we’re responding to areas where violence is already particularly high. Could a better model be strategically … looking at how you tackle poverty and inequality? We need to tackle the drivers of violence rather than just respond when there’s an incident.” (Operational professional)

Community representatives echoed these concerns in some areas, in observations that VRUs might not be getting to grips with the reality of some young people’s day-to-day life and sense of opportunity. For those who participated in this evaluation, young people’s views show positive experiences of VRU-funded interventions, but the evaluation design cannot show how these valuable insights align with the wider experience of young people.

7. Conclusions

This chapter provides conclusions.

There were no statistically significant impacts on the primary SV outcomes of hospital admissions for sharp object violent injury or homicides. However, there was a statistically significant reduction in hospital admissions for any violent injury which was a secondary outcome of interest. Since funding began, an estimated 3,220 total (or 8 per 100,000 persons) admissions had been prevented in funded areas. Analysis focusing on just violence ‘hot spots’ (within VRU areas) also indicated a potential impact on police recorded violence with injury and violence without injury offences.

The statistically significant reduction in violence without injury offences at a whole-programme level that was observed in previous years’ evaluations was not sustained. However, the wide (and overlapping with previous years) CIs for these estimates indicate a high degree of variability in the data. As such, we cannot say that the results for this outcome reflect a real change in violence.

Overall, VRUs continued to make progress in 2022 to 2023 against all elements of the whole-systems approach, albeit with, sometimes wide, variability between VRUs. Whilst overall approaches were generally considered effective, progress relative to past years was more limited, but we could argue this is reasonable to expect given that most VRUs have established their approaches and have moved to a maturing phase, where strides in progress are less common.

VRUs have continued to make good progress towards embedding an effective multi-agency approach. This included progress in engaging a wider range of partners, utilising partner expertise and resources, and coordinating multi-agency responses to violence. The introduction of the SV Duty and 3-year funding was generally viewed as an opportunity to strengthen partner buy-in, commitment and focus, and improved strategic commissioning, but there was a recognition it would take time to realise the full impacts of these. There were still areas for improvement noted in terms of meaningful engagement and shared responsibilities with all partners, and ensuring the sustainability of the multi-agency approach. Approaches to effective multi-agency working focused on strong relationships with partners, effective structures, strong leadership and having a clear vision. There was also growing support for the hybrid VRU model, which allows central coordination and effectively harnesses local structures.

There was some progress made towards effective data sharing and analysis. Progress was most evident in VRUs that had (continued to) invest in systems and access more and better quality data to improve the VRUs and partners’ understanding of those at risk. However, there was a wide range of variability across VRUs. This included many still facing challenges including access to granular data, ongoing efforts to increase data sharing and standardise data capture, and organisational restructures hindering data sharing progress. Accessing health data was still an issue for many VRUs, but it was hoped the SV Duty might help eventually overcome this. Common approaches that facilitated progress towards effective data sharing and analysis included strong relationships with partners, data governance and technological solutions, and the embedding of analysts and analytical capabilities. The former highlights the importance of embedded and effective multi-agency working as a cornerstone to implement the whole-systems approach to prevent violence.

Progress towards effective engagement of young people and communities varied greatly across the VRU. More than half of VRUs demonstrated good progress and, importantly, improved progress based on previous years where things had moved more slowly, relative to other elements of the whole-systems approach. In all VRUs, engagement appeared to be closely linked with multi-agency working, especially in relation to means of engagement and methods of communication. Where multi-agency working is showing good progress, there is often good progress with the engagement of young people and communities. Additionally, the shift to the 3-year funding commitment, in some cases, provided stability and has encouraged increased focus and commitment to young person and community engagement work. The co-design of interventions has also proven effective and popular across VRUs.

VRUs continued to make progress towards strategic and evidence-based commissioning and delivery of interventions. A total of 328 interventions were identified in the 20 VRU delivery plans. There was evidence that VRUs were more established in strategic commissioning, which was supported by the 3-year funding. However, there was still variation in the extent to which VRUs engaged partners in commissioning decisions. Although identified as priorities for this latest round of funding, and highlighted within delivery plans, professional stakeholders across the sites did not readily identify any interventions as ‘high impact’ in interviews. Those familiar with this framing from the delivery plans noted those which were high impact had followed on from earlier years’ commissioning, although some newly commissioned interventions were locally determined as high value in one area. Those which were being commissioned as a continuation had proved successful in previous years, in local assessments. This is one reason the toolkit was not key to the commissioning decisions as most running were commissioned as a continuation. Data insights were increasingly being used to inform the commissioning and targeting of interventions, but it was recognised that operationalising these and effectively engaging the identified target groups can be a challenge. There was evidence of VRU-commissioned interventions delivering on their intended early outcomes for young people, which included enhancing young people’s wellbeing, social skills, and educational and employment experience and opportunities. Yet, it was too early to say what impact interventions were having on the longer-term outcomes of reducing SV and improving community safety.

In summary, whilst progress has been mixed, and some challenges have persisted, VRUs are increasingly fulfilling their intended role to lead and coordinate the local response to violence. The 3-year funding allocations have helped with the progress made in 2022 to 2023, and the SV Duty is anticipated to further support the VRUs in their mission.

Annex A: QED method and additional outputs

This technical annex provides detail on the data sources and analytical approach for the quasi-experimental designs implemented as part of the 2022 to 2023 evaluation of Violence Reduction Units (VRUs). Additional outputs from the synthetic control group analysis are also provided.

Data sources

This section details the outcomes data accessed for the evaluation. This includes the geographical coverage and level of data, and key decisions taken when preparing data for analysis.

Hospital admissions data

Hospital admissions represent the most serious non-fatal violence, which is characterised by a violent injury that could not be treated in an emergency department. One study has found that approximately 30% of attendances to emergency departments for violent injury are admitted to hospital (Quigg et al., 2012). When a patient is admitted, their reason for admittance is recorded using the ICD-10 set of indicators. These appear in the NHS Digital records in several forms; ‘finished admissions episode’ is the most suitable way to identify these admissions. The admission is recorded against the month in which it concluded (as opposed to the admission date or the incident that led to the admission). Therefore, it may not capture the small number of prolonged admissions that may have begun but not concluded during the data collection period and some admissions may have begun before introducing funding but concluded afterwards.

In the NHS Digital data set, admissions are attributed to hospitals located within a police force area (PFA) and local authority (LA) and the month in which the admission concluded is recorded. Accordingly, admissions for violence in those hospitals are attributed to that PFA/LA, which allows the data to be used to examine differences in admission for violence in VRU and non-VRU areas and the timing of the admission closure (before or after SV funding) allows pre-post trends and a synthetic control group approach to be employed.

The analysis incorporated all hospital admissions for violent injury (ICD-10 codes: X92-Y09) and a subgroup, hospital admission for violent injury with a sharp object (ICD-10 code: X99).

Hospital admissions for any injury incurred in violence were not a primary outcome of the VRU programme, but given the range of VRU activities, it is theoretically plausible and very likely that the VRU activity could have impacted all violence. There are 16 categories of violent admissions in the ICD-10 catalogue. This includes varied mechanisms such as assault by bodily force; assault by different types of firearm; assault by drowning; assault by smoke, fire and flames; and 2 groups of ‘Other’ (specified and unspecified) assault mechanisms. We considered limiting the categories to assault mechanisms that could reasonably fit within a theoretical framework of VRU activity. However, because the range of activities by VRUs tend to aim to affect behaviour rather than the mechanism of violence, we could not confidently limit the data via theory.

We also considered using objective statistical techniques, identifying the point of maximum curvature using the data set of admissions before SV funding (2012 to 2018), to limit the data set to the most common types of violence. Using this analysis, we identified a set of 6 admission injury mechanisms, which included ‘Other unspecified’ but excluded theoretically relevant mechanisms such as handguns and rifles. This was unsatisfying as it was not theoretically sensible. In addition, such a statistical approach would create limits for the generalisability of findings beyond England and Wales. Accordingly, we opted to include all admissions for violent injury within the ICD-10 catalogue.

Homicides data

The Home Office provided homicides data at the PFA level. The data sources were:

• quarterly Homicide Index (HI) data covering fiscal Q1 2015/16 to Q2 2022/23
• monthly PRC homicides covering August 2013 to December 2022

The HI is continually updated (as police investigations and court cases progress) and is the primary data source for national publications on homicides. This data was used for the descriptive statistics and the programme-level synthetic control group analysis (using quarterly data) presented.

Recognising the low number of quarterly homicides, the evaluation team also conducted synthetic control group analysis on annual homicides. To meet the minimum number of pre-intervention periods required for the analysis, they aggregated PRC monthly homicides data to annual homicides.

To avoid biasing the impact estimates, the team removed homicides related to the following events from both data sources:

• in the financial year ending March 2017, 96 victims of the Hillsborough disaster and 4 victims of the Westminster Bridge attack
• in the financial year ending March 2018, 31 victims of the terrorist attacks that involved multiple victims, including the Manchester Arena bombing, and the London Bridge attack and 11 victims of the Shoreham air crash
• in the financial year ending March 2020, 39 victims whose bodies were found in a lorry in Essex

Police recorded crime data

The Home Office provided monthly data of police recorded violence at the Community Safety Partnership (CSP) level. Each PFA comprises multiple CSPs, which are broadly aligned to LAs. At the time of data collection, all police forces submit this data to the Home Office. Monthly counts for the following offence types were provided:

• violence with injury
• violence without injury
• possession of weapons offences

Data was also provided at the Lower Super Output Area (LSOA) level to facilitate exploratory analysis to understand the impact of SV funding in sub-areas with the highest (historical) levels of violence.

Whilst Greater Manchester Police provided data for the first time since their systems changed, the evaluators did not include these in the analysis. This was because of known under-reporting, which was clearly visible in the data and risked compromising the models.

Constructing the counterfactual

To estimate the impact of SV funding, a counterfactual (that is, what would have likely happened in the absence of funding) was constructed through:

• where available data allowed, conducting analysis at the LA or CSP-level, where treated (VRU) and untreated (non-VRU) areas are more comparable in population size and outcomes of interest
• population adjusting outcomes data to make treated (VRU) and untreated (non-VRU) areas to improve comparability
• forming synthetic control groups (a weighted average of CSPs in non-VRU areas) that followed a similar pre-SV funding trend in (population adjusted) police recorded violence to CSPs in VRU areas

The first 2 points are discussed and illustrated below using police recorded crime (PRC) data. Focusing on LA-level data and population adjustment improved the comparability of hospital admissions data. Homicides data was only available at PFA level but population adjustment was still able to improve comparability.

Analysis at CSP-level

PFAs were selected for SV funding based on levels of serious violence (SV). The 18 PFAs experiencing the highest levels of SV received SV funding and established VRUs. However, these higher levels of violence are driven by the larger populations in VRU areas.

Funded PFAs typically cover large geographical areas and/or have more densely populated urban centres (relative to non-funded PFAs). The box-and-whisker diagram below (Figure A.1) shows that the population distributions in funded (VRU) and non-funded (non-VRU) PFAs are not very comparable (that is, limited overlap between the distributions).

Figure A.1: Distribution of PFA populations (excludes outlier Metropolitan Police (population 8.9 million))

However, if smaller and more similar-sized funded and non-funded geographical Units are used (in this case, CSPs), comparability is improved. This is evident in Figure A.2 where there is greater overlap between VRU and non-VRU CSP population distributions than was observed in Figure A.1. It is still worth noting that CSP populations within funded areas were, on average, larger (reflecting a concentration of cities/large towns) than those in non-funded areas.

Figure A.2: Distribution of CSP populations

Population adjustment

Figure A.3 shows how population adjusting the outcomes can further support the construction of an appropriate counterfactual. The left-hand panel shows the average monthly count of police recorded violence with injury offences in funded and non-funded CSPs. Note that the left-hand panel very closely mirrors Figure A.2, highlighting the association between population size and outcomes of interest. The right-hand panel shows the same outcome but after population adjustment (average monthly rate per 100,000 persons), where comparability between funded and non-funded CSPs is substantially improved.

Figure A.3: Comparison of average violence and violence rates in VRU and non-VRU CSPs

Synthetic control method (SCM)

SCMs seek to construct an appropriate counterfactual by creating a synthetic control group, which is a weighted average of potential comparator areas. Comparator (non-VRU) areas (PFAs/LAs/CSPs) that are more similar to the treated (VRU) areas on pre-intervention (population adjusted) outcome trends receive a heavier weighting than those that are less similar.

As well as providing reliable impact estimates (O’Neill et al., 2020), SCMs can provide overall/average impact estimates for the entire treatment period, and average/cumulative impact estimates for each individual treatment period (for example, months, quarters). The latter is particularly useful for VRUs where it is anticipated impacts will be gradual and accumulate over time.

The precise approach was generalised synthetic control groups (Xu, 2017), which allows for the inclusion of multiple treated areas (and constructing a synthetic control group for each). Diagnostic plots were assessed to ensure there was sufficient overlap between treated and comparator areas for impact estimates to be considered reliable – this is referred to as the ‘common support’ assumption.

Consideration for the new VRUs

A key consideration for the 2022 to 2023 impact evaluation was the 2 new VRUs, Cleveland and Humberside, moving from the comparator group to the treatment group. To navigate this and provide maximum insights, the evaluators tested the following treatment specifications in the synthetic control group analysis:

• Model 1: Cleveland and Humberside initially in the comparison group but then switching to treatment group in 2022 to 2023 financial year. This model aligns to the VRU programme initiation dates in each PFA
• Model 2: Cleveland and Humberside in the comparison group only. The justification for this is that the 2 new VRUs were in a setup phase in 2022 to 2023, where direct interventions with young people and wider activity that could be expected to impact on violence outcomes were limited. This is consistent with the VRU Impact Feasibility Report, which recommended only testing for impacts on the original 18 VRUs after 18 months of programme delivery to allow sufficient time for the anticipated longer-term preventative effects of VRUs to materialise (MacLeod, 2020)
• Model 3: Cleveland and Humberside excluded from the data to test for any potential contamination of estimates resulting from either of the above models

Assumptions and limitations

A key consideration for the synthetic control group analysis was the impact of COVID-19 on violence. Specifically, the impact of lockdowns/restrictions on violence outcomes. It was theorised that restrictions on people’s movement reduced the opportunity for violence. To test this theory, Google mobility data was examined.

Figure A.4 shows that on average the LAs in VRU and non-VRU areas followed a very similar trend in residential movement (time spent at home) between February 2020 and December 2021. Regression analysis, which tested the interaction between VRU status and time on movement, confirmed there was no statistically significant difference between LAs in VRU and non-VRU areas (estimate = 0.00009, p-value = 0.89).

Figure A.4: Percentage change in residential movement in VRU and non-VRU areas

Recognising the above, the impact estimates (which are the differences between VRU areas and the synthetic control group constructed from non-VRU areas) can be considered net of any effects on violence resulting from changes in movement related to COVID-19. In other words, it was not necessary to adjust the synthetic control group analysis for changes in movement because VRU and non-VRU areas experienced very similar (and not statistically significantly different) trends in movement. However, it will be important for police recorded violence trends (and impact estimates) to be continually monitored as more data not affected by COVID-19 restrictions become available.

Whilst the above showed the differences in movement between LAs in VRU and non-VRU areas were, on average, similarly affected by COVID-19, Google mobility data was not available at more granular levels to test this at the LSOA level. As such, some caution is advised with the analysis of ‘hot spots’, which used LSOA-level PRC data. It is possible that COVID-19 affected movement differently in LSOAs.

Another consideration was the level of analysis conducted to understand changes in violence trends. Whilst monthly data was provided, the main analysis presented included aggregating the data to fiscal quarters. The team took this decision to better capture underlying trends in violence, which were less clear (owing to volatility) in the monthly data.

It is important to note the 2 main potential limitations of the police recorded violence data:

• not all violence is reported to the police by the public and this is not captured in the data. Similarly, the police do not detect all violence. Increased levels of reporting and/or detection do not necessarily reflect a real increase in the levels of violence an area is experiencing. It is possible that VRU activity (for example, building relationships with communities affected by violence, better use of data) and Grip activity (that is, additional enforcement) could affect the reporting and/or detection of violence
• changes in how the police record violence. This can include changes to and/or improvements over time in data capture systems

Whilst results from the police recorded violence data should be interpreted with the above points in mind, analysis was conducted at the whole-programme level, which can be considered less affected by potential changes in the reporting, detecting or recording of violence in specific PFAs (they are averaged out against other PFAs).

There are some limitations to using hospital admissions data as an indicator of SV in a PFA. Firstly, the catchment area of emergency departments are not necessarily co-terminus with PFAs. As patients typically travel to their closest emergency department for treatment (Haas et al., 2015), the hospital in which they were treated may not be in the PFA where the violent incident occurred. Similarly, patients injured in more rural areas – which are also likely to have lower levels of violence and therefore less likely to receive SV funding – have to travel to population centres such as towns and cities for treatment meaning that violence captured by the hospital(s) in a PFA may not have occurred in that area.

Hospital admissions are also susceptible to resourcing issues that may affect their validity. For example, hospitals with fewer resources and staff may be less likely to admit a patient for a violent injury than a hospital with more resources. In addition, the availability of a minor injury treatment centre within the hospital’s catchment area may reduce the number of patients attending (Rudge et al., 2013).

Additional results

The table below shows the estimates and inference across all outcomes and model specifications.

Model	Outcomes	Estimate	Std. Error	Lower CI	Upper CI	P-value
Model 1	Homicides	-1.54	2.35	-6.35	2.70	0.41
Model 2	Homicides	-1.53	2.46	-6.35	3.05	0.51
Model 3	Homicides	-1.59	2.43	-6.42	2.80	0.41
Model 1	Hospital admissions sharp object	-0.78	0.57	-1.75	0.39	0.20
Model 2	Hospital admissions sharp object	-0.90	0.57	-2.14	0.18	0.09
Model 3	Hospital admissions sharp object	-0.70	0.56	-1.74	0.42	0.25
Model 1	Hospital admissions any violence	-9.50	3.69	-16.22	-1.95	0.01
Model 2	Hospital admissions any violence	-7.96	3.60	-13.99	0.00	0.05
Model 3	Hospital admissions any violence	-9.19	3.63	-15.58	-1.38	0.01
Model 1	PRC: Violence with injury	59.83	59.53	-124.24	100.49	0.89
Model 1	PRC: Violence without injury	61.51	230.29	-398.84	496.84	0.82
Model 1	PRC: Possession of weapon offences	3.89	16.79	-34.93	32.32	0.87
Model 2	PRC: Violence with injury	6.89	54.67	-146.84	64.91	0.56
Model 2	PRC: Violence without injury	-62.95	190.25	-556.57	219.24	0.42
Model 2	PRC: Possession of weapon offences	-9.45	20.96	-62.50	23.71	0.38
Model 3	PRC: Violence with injury	61.72	63.87	-131.56	119.02	0.80
Model 3	PRC: Violence without injury	63.79	224.27	-339.11	500.96	0.76
Model 3	PRC: Possession of weapon offences	4.33	17.55	-39.53	33.13	0.91

The figures below show the synthetic control group analysis outputs for PRC outcomes.

Figure A.5: VRU and synthetic control group trends – Violence with injury – Model 2

Figure A.6: VRU and synthetic control group trends – Violence without injury – Model 2

Figure A.7: VRU and synthetic control group trends – Possession of weapons – Model 2

Meta-regression

The evaluators tested the following models for variation between VRUs.

Model	Hospital admissions for any violence	Hospital admissions by sharp object
	P-value	R squared	Significant	P-value	R squared	Significant
Model 1: Total funding	0.1	15.53		0.13	9.62
Model 2: Funding per capita	0.05	19.4	*	<0.01	55.04	*
Model 3: Spending on interventions	0.09	16.47		0.17	6.23
Model 4: Spending on interventions per capita	0.03	26.57	*	<0.01	55.08	*
Model 5: Spending on high impact	0.01	39.96	*	0.07	14.97
Model 6:Proportion of intervention spending on high impact	0.07	12.73		0.5	<0.01

Notes:

1. * indicates statistical significance.

Estimating the impact of VRUs (Independent of Grip activity)

The Home Office provided monthly police recorded offending data at the LSOA level for 33 PFAs. After data cleaning and sense checks, the team retained data for 16 PFAs for analysis. Sense checks included comparing PFA violence count totals from the LSOA-level data, where recording is more variable, to CSP-level data to understand data coverage and consistency over time. The table below shows the 16 PFAs and their SV funding (or ‘treatment’) status.

SV funded	Not funded
Avon & Somerset	Cambridgeshire
Bedfordshire	Dorset
Hampshire	Durham
Merseyside	Hertfordshire
Metropolitan Police	Lincolnshire
Northumbria	Northamptonshire
Nottinghamshire	Surrey
Sussex	Wiltshire

Data was aggregated to quarterly (from monthly) to better capture underlying trends and facilitate the synthetic control group approach (see next section ‘Defining hotspots’).

Defining hotspots

LSOAs in SV-funded PFAs were defined as hotpots based on the total violence (with and without injury) in the 3 years prior to SV funding. LSOAs were arranged (high to low) by their total violence and assigned to quantiles (in equally sized groups). Two definitions were operationalised and tested:

• LSOAs in the top decile (10%) of total violence, within each PFA
• LSOAs in the top centile (5%) of total violence, within each PFA

Recognising that PFAs cover large geographical areas and the focusing of Grip activity, attribution is likely stronger for analysis focused on the top 5% of LSOAs. The table below shows how these 5% of LSOAs accounted for approximately a quarter of all violence with injury offences in their wider PFA (based on year ending March 2019 data).

Police force	Total offences	Offences in hotspots	Offences in hotspots (%)
Avon & Somerset	12,409	3,456	27.9
Bedfordshire	4,685	1,467	31.3
Hampshire	20,609	5,001	24.3
Merseyside	14,602	3,849	26.4
Metropolitan Police	77,038	18,355	23.8
Northumbria	13,275	3,661	27.6
Nottinghamshire	13,662	3,455	25.3
Sussex	13,871	3,928	28.3

Results

There were statistically significant reductions in violence with injury and violence without injury offences after introducing SV funding in violence hotspots. The table below shows that in the top 5% of LSOAs, there was an average (per LSOA) reduction of 32 violence with injury and 17 violence without injury offences over the SV funding period (April 2019 to December 2021), relative to the synthetic control group. The total reduction (across the 541 LSOAs defined as hotspots) was 17,424 violence with injury and 9,351 violence without injury offences.

Hotspot definition	Offence	Cumulative impact	Lower CI	Upper CI
Top 10%	Violence with injury	-20.18	-23.82	-16.60
	Violence without injury	-4.48	-12.95	2.51
Top 5%	Violence with injury	-32.21	-39.15	-25.73
	Violence without injury	-17.28	-32.46	-2.89

Figure A.8 shows the trends in violence with injury and violence without injury offences for the SV-funded hotspots (both definitions) against the synthetic control groups. In all cases, trends in SV-funded hotspots started to diverge from the synthetic control groups (indicating a reduction) shortly after introducing funding and this was generally maintained until the first quarter of the financial year ending March 2022. Trends in more recent periods were more similar.

There were substantial drops in violence during the first, third and fourth quarters of the financial year ending March 2021 in SV-funded hotspots and the synthetic control groups, which align with COVID-19 national lockdowns. For violence with injury offences, SV-funded hotspots appeared to have greater reductions during the national lockdowns, relative to the synthetic control groups. Whilst this could be the result of Grip/Surge activity, it is also possible that SV-funded hotspots were more affected by the national lockdowns. It is worth noting that Surge/Grip activity had a specific focus on managing/preventing anticipated spikes in violence following lockdowns.

Figure A.8: Violence trends in SV-funded hotspots and synthetic control groups

Annex B: Case studies

The evaluation team conducted 3 ethnographic case studies, including observations and filming/photography, on one/2 Violence Reduction Unit (VRU)-funded interventions in a local area. This annex summarises these, with thematic films also produced across the set of 3. The team chose the sites through a process of engagement and discussion with VRU Leads and individual interventions. Half of the sites expressed interest, with 5 initially able to meet the timelines for ethnography and 3 then selected for geographic diversity and a cross-section of intervention type. All staff and volunteers, young people and parents/carers were provided with advance information about the ethnography, and told it was part of a study for the Home Office, which funds the intervention.

The team gathered consent forms from young people and parents/carers where required for all young people who participated in the ethnography, including clarification as to whether they gave consent for their image to be gathered and shared as part of the ethnography and written/film outputs.

The individual young person/people who were then followed at each site for a day as part of the ethnography were selected by the interventions, using information sheets and consent forms provided by the ethnography/evaluation team in advance. The day was selected in discussion with the young person, and they were asked again for their consent on the day, with repeat clarification that they could opt out at any time. They selected how the day should be spent and where the ethnographers accompanied them. The ethnography/evaluation team’s safeguarding and disclosure protocols were used throughout the approach to and conduct of the ethnography.

Case study 1

Situated within a suburban neighbourhood, a local Pupil Referral Unit (PRU)^{[footnote 15]} is using VRU funding as part of their larger strategic initiative to focus on a rewards-based, positive encouragement culture to increase attendance and confidence, and reduce violent and anti-social behaviour. The school had a positive atmosphere with little disruption observed throughout the entire school day. Staff encouraged young people to ‘step away’ or ‘remove themselves’ from a situation when they were feeling agitated. This meant young people were often in the hallways with a light touch approach from staff to emotionally support them to reintegrate back into the classroom. Student artwork covered the halls, which the school noted the young people were proud of and had reduced destruction of property, and given a sense of ownership and pride over the space.

With attendance being one of the largest barriers for this community, VRU funding was used to pay for regular outings, such as base-jumping or adventure activities, to incentivise young people to come into school. Teachers noted these small trips are the only days they have full attendance of students, and students equally commented on these activities being the highlight of their time.

This PRU has partnered with a local community boxing club within walking distance of the school to provide weekly combat and boxing classes for students. Whilst the funding provides free lessons during school hours, the owners subsidise some of the young people to come in the evening and continue training. This programme is extremely popular and has led to a major cultural shift within the school, with students eager and engaged in the programme. Teachers and leadership noted the discipline and physicality helped channel some of the emotional struggles and release some energy.

The gym is a family business with a local husband and wife team and their daughters competing at national level. They are passionate about their community and about the impacts of combat training on people.

“Not just this gym, all gyms are helping change lives every day. There is something different from combat sports that makes it stand alone from other projects. It demands that we look at ourselves and look truthfully. They demand respect, both self-respect and respect for others. They demand hard work, dedication, commitment and sacrifice. This plays out for their rest of their lives.” (Owner)

When entering the gym, the young people were quick to smile and banter with the owners as they prepared for the session. There was a strong sense of familiarity, safety and joy in the space. Some young people joined the owners in the kitchen to cook and prepare homemade burgers whilst others took off their shoes and got ready to train. The owners explained they provide meals and cooking lessons for the young people to learn healthy habits and nutrition, as well as noting many would go hungry after school. The class was mixed age and gender, with everyone collaborating and working hard whilst also laughing and enjoying themselves.

A teacher pointed out a confident teenage girl laughing as she trained, who when she arrived 6 months ago “would not speak or crack a smile. The only time we heard her was when her rage took over and she would scream and shout. Now look at her!” She is proud of where she is now and the support the PRU and these extracurricular activities have had on her. She intends to gain her beauty qualification and move back to her family’s home country to become a hairdresser. She explained her own journey and transformation:

“I got kicked out for having a fight with a girl that was racist to my little brother…I broke her nose. I feel like all the programmes that they [the PRU] do are there to help us learn more and like, be better. Like they do all this stuff to help with our behaviour. I started doing combat because I went with the school and really enjoyed it so started doing it after school as well.

“I like it because of the support I get from all the staff members. They push us beyond our limit. And they teach us how to keep our anger in and then let it out on a punching bag later. Like keep it in all day until I go to combat.”

Ethnographers spent a full day in the company of a 17-year-old student, identified as a success story within the school and combat program. He has been a student at this PRU since 12 years old, yoyo’ing back to the PRU by choice, feeling mainstream education and large class sizes too overwhelming and upsetting. This young person was originally removed from mainstream school due to fighting and violent behaviour towards teachers, something that is hard to imagine now. Today he is mild-mannered, polite, focused and giving – much of which he attributes to finding boxing. He was observed being a quiet leader within the school, creating an atmosphere of respect and good behaviour. His physically large presence and the knowledge that he was skilled at sport gave him social capital and an air of respect from his fellow students. He spoke about his ‘bad behaviour’ in the past and felt he had been leading down a path that would end in severe violence. When discussing his home life, local community and where he hung out, he spoke about the lack of space at home with 5 people in a 2-bedroom house but also avoiding many public spaces fearing violent confrontation. His girlfriend and the gym were safe spaces in his life. In his own words:

“Nowhere is really safe but sometimes you feel a little more safe. I got robbed by 6 people here about 2 years ago. They took my old rusty bike… 3 or 4 of them had machetes with them, so it was pretty scary.”

Passionate about training, he has now got into college to study to gain a qualification in coaching. He will also start working as a coach for young people when he turns 18 – something he is looking forward to. Whilst at the gym, he mentored and supported younger boxers whilst also training with professional and semi-professional boxers. Everyone spoke of B as someone they were extremely proud of for his hard work, attitude and quiet leadership.

Leadership, teachers and coaches all identified how the funding has positively impacted individual young people’s lives, providing them with incentives for attendance at school, as well as discipline and a safe outlet for their emotions. The combat-fighting gym is a central part of the community and a safe space in an environment that feels otherwise unsafe. The owners are pillars of the community and recognisable throughout town, garnering respect as they walk down the street (ethnographers even observed someone hopping out of their car to come shake their hands). The settings also provide peer-to-peer leadership and role models that impacted the overall culture of the PRU. Leadership noted this cultural shift with new young people arriving at the school and ‘feeling the atmosphere’ of support and positive opportunities.

Case study 2

The second case study, a mobile youth centre, implements a range of VRU-funded interventions to engage young people in the community. These interventions include leadership training, a youth club, 1-to-1 mentoring sessions, and a youth café where young people cook for their peers. The activities provided by the centre serve as a platform for personal growth, skill development and social interaction among young people. Through these interventions, young people find valuable opportunities to explore their interests, learn new skills and connect with their peers. It caters to a range of ages, as the VRU-funded interventions are designed to be inclusive and accessible to young people of various age groups, from 14 (Year 10) through to 18 (Year 13/sixth form).

The centre aims to engage a wide range of young people with its activities and has a flexible approach to implement their VRU-funded interventions. The initiative uses different local facilities, including their own building as well as partnering with local charities and organisations. The flexibility of the activities, such as football practice being held in 2 different spaces, allows for broader engagement and a wider impact on young people within the community. Alongside the flexibility of its locations, the centre tailors the interventions to young people’s specific needs and interests, incorporating activities such as leadership training, social hangouts, arts and crafts, and podcasting. By offering a diverse range of interventions, the centre considers it can meet the varied preferences and engagement styles of the young people they aim to support, and this was demonstrated in the observations.

The centre works to align individual activities with the longer-term goals of the VRU overall. Its focus on personal growth, development and long-term planning echoes VRU objectives to reduce violence, promote positive youth engagement, and foster young people’s sense of purpose and direction. The leadership training, for instance, aims to equip young people with essential skills and qualities to positively impact their lives beyond the intervention.

The centre actively engages its young people in co-production and co-design, creating a sense of ownership and agency. They have a group of ‘elected youth’ who run a board that regularly meets every school term to discuss the running of the centre and provide suggestions and feedback for future activities. The input from the youth board is valued and incorporated into the organisation’s decision-making processes. For example, introducing a ‘chill space’ where young people could spend time informally with each other was a successful initiative that began with ideas from the board. In addition, cooking sessions where young people cook for others was another initiative suggested by the youth board. This participatory approach shows the centre as responsive to the needs, perspectives and aspirations of young people themselves.

The staff demonstrated a clear understanding of the long-term aim and impacts of the sessions and interventions. Many of the staff and volunteers are from the community they are supporting or used the centre as young people themselves. As such, they recognise the importance of providing a safe space for young people to avoid not only risky situations, but also the importance of providing opportunities for young people to experience new things and learn more about themselves and their community. They strategically designed the sessions and interventions to offer young people meaningful experiences, transferable skills and opportunities for building self-confidence.

There are clear indications of potentially-lasting impacts resulting from the VRU-funded interventions provided by the centre. The involvement of young people in the leadership board, the authentic friendships formed, and the skills and experiences gained through their engagement with the centre contribute to their personal growth and resilience. The observations and conversations showed that these can have a lasting impact on their trajectories, empowering them to pursue their goals and become active contributors to their communities through the centre.

The work of the centre demonstrates value in supporting and empowering young people. It also demonstrates the benefits of an iterative approach, whereby interventions are tailored and grow in response to the needs of young people. By providing a safe and inclusive space, offering a variety of interventions tailored to their interests, and actively involving them in decision-making processes, the centre fosters a sense of belonging and purpose among young people. The positive feedback from the young people themselves, the engagement of the youth board, and the observed personal growth and transformations all highlight the significant value of the VRU-funded interventions in enabling young people to thrive and make positive contributions to society. In the words of one of the young people involved in the centre:

“[It] helped me discover my potential and gave me a sense of purpose. I am driven to make a difference, and with the skills and support I have gained, I have the power to do so. It’s not just an intervention; it’s a springboard for my dreams and aspirations.”

Case study 3

The third case study is a community youth centre that hosts a range of VRU-funded interventions in a semi-urban area. The VRU-funded interventions include boxing, football, climbing, arts and crafts, and cycling activities for young people aged 8 to 16 years old. In addition, the centre receives other funding to provide hot meals for young people, particularly after physically demanding activities.

The centre serves as a safe space for young people to interact and engage in activities that promote personal growth, skill development and social interaction. The range of interventions offered ensures inclusivity and accessibility for young people of different age groups. Although the centre is open to those up to the age of 16, it tends to attract more young people between the ages of 8 and 13.

The centre operates across different facilities. The primary location is the community centre itself, which has a history of community development programs and interventions. The centre also uses local facilities for various activities. These locations are easily accessible and well-known to the young people in the community.

The centre’s interventions align with the longer-term goals of the VRU, which include reducing violence, promoting positive youth engagement, and fostering young people’s sense of purpose and direction. The activities at the centre aim to equip young people with essential skills and qualities that positively impact their lives beyond the intervention.

The centre actively engages young people in co-production and co-design. The centre allows young people to suggest activities, and the staff responds by incorporating those suggestions into the schedule. The input from young people is valued and considered in the organisation’s decision-making processes. This participatory approach demonstrates the centre’s responsiveness to the needs, perspectives and aspirations of young people themselves.

The staff and coaches at the centre, many of whom are from the surrounding community, demonstrate a clear understanding of the long-term intention and impacts of the sessions and interventions. They have first-hand knowledge and experience of the needs of young people in the community. The coaching and mentoring provided by the staff aim to develop empathy, responsibility, discipline and care for others among the young people.

The VRU-funded interventions have clear indications of potentially longer-lasting impacts for individual young people and beyond. There were examples given of the centre ‘transforming the lives’ of individual young people and having a positive influence on whole families, including parents. For example, one child’s involvement in the boxing program led his father to become a coach and community mentor. This has created a network of support within the community and increased engagement with the centre.

The work of the centre demonstrates the value of supporting and empowering young people. By providing a safe and inclusive space, offering a variety of interventions tailored to their interests, and actively involving them in decision-making processes, the centre fosters a sense of belonging and purpose among young people. The positive feedback from young people, the engagement of parents and other service providers, and the observed personal growth and transformations all highlight the significant value of the VRU-funded interventions in enabling young people to thrive and make positive contributions to society.

In conclusion, the youth centre serves as a valuable resource for young people in the community. Through a range of VRU-funded interventions, the centre provides opportunities for personal growth, skill development and social interaction. By aligning its activities with the longer-term goals of the VRU and actively involving young people in co-production and co-design, the centre empowers young people and fosters a sense of belonging and purpose. The positive impacts observed in individuals and families demonstrate the significance of the VRU-funded interventions in enabling young people to thrive and make positive contributions to their communities.

References

Braga, A.A., Turchan, B.S., Papachristos, A.V. and Hureau, D.M. (2019) ‘Hot spots policing and crime reduction: An update of an ongoing systematic review and meta-analysis’. Journal of Experimental Criminology, vol. 15, pp.289 to 311.

Gaffney, H., Farrington, D. and White, H. (2021a) Cognitive Behavioural Therapy. Available from: https://youthendowmentfund.org.uk/wp-content/uploads/2021/06/CBT-Technical-Report-.pdf [accessed 23/11/2023].

Gaffney, H., Jolliffe, D. and White, H. (2021b) Emergency department violence interventions. Available from: https://youthendowmentfund.org.uk/wp-content/uploads/2021/12/Emergency-Department-Violence-Interventions-Technical-Report-1.pdf [accessed 23/11/2023].

MacLeod, S., Cutmore, M., Deoudes, P. and Brennan, I. (2020) Violence Reduction Units – Impact evaluation feasibility study: Summary report for VRUs and wider stakeholders. Research Report 117, Home Office, London. Available from: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/910823/violence-reduction-units-impact-evaluation-feasibility-study-horr117.pdf [accessed 18/11/2023]

NHS Digital (2014) ISB1594: Information Sharing to Tackle Violence Minimum Dataset. Available from: https://digital.nhs.uk/data-and-information/information-standards/information-standards-and-data-collections-including-extractions/publications-and-notifications/standards-and-collections/isb1594-information-sharing-to-tackle-violence-minimum-dataset [accessed 22/11/2023]

O’Neill, S., Kreif, N., Sutton, M. and Grieve, R. (2020) ‘A comparison of methods for health policy evaluation with controlled pre-post designs’. Health Services Research, vol. 55(2), pp.328 to 338. Available from: https://onlinelibrary.wiley.com/doi/10.1111/1475-6773.13274 [accessed 23/11/2023].

ONS (2021) Police recorded offences involving knives or sharp instruments: methodology changes. Office for National Statistics. Available from: https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/methodologies/policerecordedoffencesinvolvingknivesorsharpinstrumentsmethodologychanges [accessed 16/11/2023]

PHE (2019) A whole-system multi-agency approach to serious violence prevention: A resource for local system leaders in England. Public Health England, London. Available from: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/862794/multi-agency_approach_to_serious_violence_prevention.pdf [accessed 16/11/2023]

Quigg, Z., Hughes, K. and Bellis, M.A. (2012) ‘Data sharing for prevention: a case study in the development of comprehensive emergency department injury surveillance system ad its use in preventing violence and alcohol-related harms’. Injury Prevention, vol. 18(5), pp.315 to 320. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3461757/ [accessed 23/11/2023].

Rudge, G.M., Mohammed, M.A., Fillingham, S.C., Girling, A., Sidhu, K. and Stevens, A.J. (2013) ‘The combined influence of distance and neighbourhood deprivation on emergency department attendance in a large English population: a retrospective database study’. PloS one, vol. 8(7), e67943. Available from: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0067943 [accessed 23/11/2023].

Smith, V. & Wynne-McHardy, E. (2019) An analysis of indicators of serious violence: Findings from the Millennium Cohort Study and the Environmental Risk (E-Risk) Longitudinal Twin Study. Research Report 110.

An analysis of indicators of serious violence, Home Office, London. Available from: https://assets.publishing.service.gov.uk/media/5d36d7fa40f0b604d8e5fe4b/analysis-of-indicators-of-serious-violence-horr110.pdf [accessed 22/11/2023]

Xu, Y. (2017) ‘Generalized synthetic control method: Causal inference with interactive fixed effects models’. Political Analysis, vol. 25(1), pp.57 to 76. Available from: https://www.researchgate.net/publication/313876970_Generalized_Synthetic_Control_Method_Causal_Inference_with_Interactive_Fixed_Effects_Models [accessed 23/11/2023]

Footnotes

1. The justification for the inclusion of any violent injury was that given the range of VRU activities, it is theoretically plausible and very likely that the VRU activity could have had an impact on this. Further detail is provided in Annex A. ↩

2. The Serious Violence Duty was introduced in January 2023 and placed a requirement on specified authorities for a local government area to work together and plan to prevent and reduce serious violence. For more information, see Serious Violence Duty - GOV.UK ↩

3. Note that in September 2021, Cleveland and Humberside were awarded bespoke funding to address increases in SV because they were the next 2 highest forces for hospital admissions for injury with a sharp object, and their volumes had risen during the pandemic. The purpose of the funding was to implement hot spot policing in the highest SV hot spots in Cleveland and Humberside. ↩

4. See PHE (2019). ↩

5. At the time of writing (early 2023), ONS publications for crime over time also did not include Greater Manchester for this reason. ↩

6. SV areas: Avon & Somerset, Bedfordshire, Hampshire, Merseyside, Metropolitan Police, Northumbria, Nottinghamshire, Sussex. Non-SV areas: Cambridgeshire, Dorset, Durham, Hertfordshire, Lincolnshire, Northamptonshire, Surrey, Wiltshire. ↩

7. Note that LA representatives accounted for 28 of the 107 responses to the survey. ↩

8. A day of engagement workshops “Finding solutions to a fairer society and safer communities” with young people – organised and delivered in conjunction with the Hope Collective. ↩

9. See youthendowmentfund.org.uk/toolkit/ ↩

10. See youthendowmentfund.org.uk/toolkit/ ↩

11. See youthendowmentfund.org.uk/toolkit/ ↩

12. See: youthendowmentfund.org.uk/toolkit-what-works-to-prevent-youth-violence/ ↩

13. The delivery plan template asks VRUs if an intervention is in the YEF Toolkit, what the estimated impact is on violent crime of the intervention and “If no, or if you believe the estimated impact to be higher than the YEF’s estimation, what impact have you assessed the intervention to have? Please state what evidence/resources you have used to assess the evidence base of the intervention.” ↩

14. Pupil Referral Units are a key alternative education provision for pupils who cannot attend mainstream school, for reasons including the nature or level of their support needs or being excluded from mainstream school. ↩

15. Pupil Referral Units are a key alternative education provision for pupils who cannot attend mainstream school, for reasons including the nature or level of their support needs or being excluded from mainstream school. ↩