Supporting the youngest children in the youth justice system: what works to reduce offending and improve outcomes?

Decorative thumbnail
This report summarises findings from a structured review of the evidence for alternatives to CJS responses, which was undertaken to establish what works to reduce offending and improve outcomes for children aged 8-13 years in contact with the criminal justice system.

About this report

This report was commissioned by the Local Government Association and produced by Dr Patricia J. Lucas and Dr Jo Staines, independent social researchers. The recommendations within the report are those of the authors, for the consideration of all those engaged in supporting younger children in the youth justice system.

Acknowledgements

Our thanks to colleagues in the Youth Justice Board for assistance with evidence gathering and contact with Youth Offending Services across the country. Our thanks to the individuals who responded to our call for evidence, and to those who provided further assistance and information to us (Alyce Simpson, Andrea Underhill, Andrew Griffiths, Annette Irvine, Cathy James, Christa Matthews, Claire Bowers, Emily Gray, Frederick VanHeerden, Jethro Bogdanov, Julie Cummins, Kim Stephenson, Liza Kinnane, Lloyd McDonald, Louise Franklin, Lucy Frampton, Lyndon Lewis, Lyndon Lewis, Marie McLaughlin, Mark Phillips, Mike Rees, Natasha Mcdonald, Nicola Suttenwood, Rachel Roberts, Sally Chugg, Sarah Dubberley, Sarah Forrest, Sheila Whitehead, Simone Fox, Steph Webber, Tanya Gillett, Vivian Blake, Williams, Claire). Thanks to Lee Clay for assistance with literature screening, and to Flora Wilkie for support and guidance.

Introduction to language used in this document

Given there is debate about whether harmful acts by children can or should be considered crimes, it follows that language for harms caused by children is itself controversial. Offences committed by children fall along a continuum from relatively minor harms to very serious criminal behaviours. We recognise that the cumulative impact of repeated harms can be distressing for victims, but we try to avoid the language of crimes in this report. We distinguish between those more minor incidents and more serious offences by referring to the former as ‘harms’, particularly when perpetrated by a child under the age of criminal responsibility and using ‘offences’ for those incidents that are formally recognised and/or prosecuted as crimes (including where out of court actions are used to avoid a conviction).

When considering the international research literature, questions of appropriate language are compounded by cultural and contextual differences. For example, what might be described within the English education system as emotional and behaviour difficulties, may be described in a clinical context as Conduct Disorder (CD). What may be described in North American literature as delinquency, might be described in the UK as offending behaviour. When reporting findings from previous research in this review, we preserve the language, and therefore the meanings, used by the original authors.

Unless specified, throughout the report, we use the term ‘younger children’ to refer to those aged 8 – 13 years; where research or approaches relate to those aged under 8, we use ‘very young children’, and ‘older children’ to refer to those aged 14 years or over.


You will find any relevant references within the text followed by a number in brackets. You can find the full source in the references section.

1. Executive summary

At 10 years old, the Minimum Age of Criminal Responsibility (MACR) in England, Wales and Northern Ireland is one of the youngest in the world. There is a strong argument for increasing the MACR to 14 years. Research suggests 10 years is developmentally inappropriate. It also contravenes the United Nations Convention on the Rights of the Child (UNCRC) and is harmful; early contact with the criminal justice system (CJS) increases the likelihood of future crimes. Few children aged under 14 years commit crimes, and their offences are largely minor.

This report summarises findings from a structured review of the evidence for alternatives to CJS responses, which was undertaken to establish what works to reduce offending and improve outcomes for children aged 8-13 years in contact with the criminal justice system.

There is a paucity of research specific to younger children in contact with the criminal justice system. Few intervention programmes were found in either the international research literature or in UK practice that addressed the particular needs of the 8-13 year old age group.

Evidence from approaches used with a wider range of children does suggest a number of effective approaches:

  • There is strong evidence that family systems approaches (e.g. family therapy and Multi Systemic Therapy) are effective at reducing disruptive behaviour, antisocial behaviours, Conduct Disorder and substance use for all ages of children and some evidence they are more effective in younger children. These approaches are being used in practice in the UK.
  • Behavioural approaches (e.g. behavioural contracts) have been shown to be effective for reducing persistent offending behaviour for all ages of children.
  • Police-led diversion schemes, using cautions, restorative actions and referrals to other services to avoid arrest, have been shown to reduce re-offending rates by about 20 per cent relative to usual practice. However, we also know that contact with the criminal justice system generally increases re-offending rates. Therefore, care should be taken when using police-led approaches not to increase contact with offending communities or negative contacts with the police.
  • While there is a strong evidence base to suggest restorative justice approaches are effective at both reducing repeat offending and improving victim satisfaction rates in older children, there is a lack of evidence about effectiveness when used with children under the age of 14 years. Restorative justice may be less appropriate at earlier developmental stages.

In current UK practice, Youth Offending Services (YOS) and Youth Offending Teams (YOT) describe working in individualised ways taking each child’s context into consideration. However, beyond this, very few programmes were found that address the specific needs of this age group. Where innovative practice exists, we do not yet have evidence from robust evaluations to support clear recommendations.

Victims of crimes do not report greater satisfaction when children receive criminal sentences. Victims feel most supported when their voices are heard and the harm to them is taken seriously, regardless of the outcome for the offender.

Based on the available evidence, the authors of this report therefore recommend:

  • Raising the age of criminal responsibility to 14 and transferring responsibility for children aged 10-13 years who cause harm from youth justice to children’s social care.
  • Expand access to systemic therapies, cognitive or behavioural therapies for younger children whose behaviour causes harm.
  • Build a system that recognises and responds to the impact of Adverse Childhood Experiences and Trauma on child behaviour. This is likely to mean providing training for professionals across sectors (including the police). It may also mean introducing earlier assessments to identify children likely to need more intensive support.
  • Avoid using cautions, court orders, or convictions as necessary qualifiers for intervention (for example from mental health teams). It is not appropriate for a criminal justice response to be required before young children receive the support they need. Needs should be assessed, and high-quality intervention considered at the earliest signs of aggressive, antisocial or disruptive behaviour.
  • Where used, police-led diversion should occur pre-arrest, to avoid unnecessary criminalisation. Diversion should be embedded within a strategy which seeks to reduce punitive contacts with the police, and supports the police to link children to other supporting services (e.g. referrals to Child and Adolescent Mental Health Services (CAMHS) or local youth activities).
  • Take a youth-led approach to designing or selecting activities and interventions. Evidence from practice suggests this will increase uptake and avoid wasting resources on unpopular programmes.
  • Neither area-based curfews, nor any scheme which brings younger children into contact with older children who are more entrenched in their offending should be used. Neither are supported by the evidence and are unlikely to be a good use of resources.
  • Develop a culture of cooperation and support between linked YOS. In case studies, participants reported this facilitated programme success regardless of the approach taken. Good relationships between services improved the experiences of organisations, professionals, families and children.

There are also gaps in our knowledge that need to be filled. We recommend:

  • Recording and monitoring non-criminal justice or diversionary actions, these should be recorded anonymously and reported at aggregate level to avoid appearing on individual children’s records. This would enable monitoring and evaluation of the use of tools such as youth justice orders, injunctions to prevent nuisance and annoyance and out of court disposals.
  • Consider the differential impact of crime reduction interventions according to child age. This is needed both when reporting on the impact of commissioned activities and when commissioning research.
  • Develop and test interventions targeting children aged 8-13 years engaged in offending behaviour.
  • Commission research into models of restorative justice or mediation appropriate for younger children, including when victims and offenders are in close relationships.

2. Context

2.1 The age of criminal responsibility in children in the UK

The Minimum Age of Criminal Responsibility (MACR) in England, Wales and Northern Ireland has not changed in almost 60 years, and at 10 years is one of the youngest in the world. In Scotland, the age was recently increased to 12 years with a plan to rise to 14 years, and the Government of Wales has indicated their support for the same.

Neuro-psychological evidence suggests that most children at age 10 will not yet have the developmental maturity and foresight for criminal responsibility to be appropriate.

The current MACR is in contravention of the United Nations Convention of the Rights of the Child. It also results in inconsistencies within UK law, where children of the same age are understood very differently in different legal contexts.

Evidence does not suggest that contact with the criminal justice system (CJS) is an effective way to reduce crime. Instead, the CJS seems to be criminogenic; exposure to criminalised contexts increases the risk of reoffending.

Find out more on the MACR

The minimum age of criminal responsibility (MACR) in England, Wales and Northern Ireland was set at 10 years in 1963. This is considerably younger than many countries across the globe, and there have been frequent calls for an increase in the MACR in England and Wales on the basis that it contravenes children’s rights [1], and because of progress in our understanding of child development, and competencies in other areas of law. It is also not uniform across the UK. Scotland has already increased the MACR to 12 [2], and is reportedly considering a further increase to 14. The Welsh Government has already indicated support for increasing the MACR in Wales, although it currently does not have the power to do so [3]. A review of the Youth Justice System (YJS) in Northern Island also recommended an increase to 12 years, potentially rising to 14 [4].

Neuro-psychological evidence suggests that most children at age 10 will not yet have the developmental maturity and foresight for criminal responsibility to be appropriate. We now understand that children do not reach full neurological maturity until early adulthood. Research demonstrates significant variation between children in the development of skills including reasoning, executive functioning, logical thinking, problem-solving and empathy [5-7]. Because of this, younger children are unlikely to be able to fully participate in criminal trials [8].

The current MACR is in contravention of the United Nations Convention of the Rights of the Child, which recommends an age of no less than 12 and an ‘ideal’ age of 14 [9]. Age of legal responsibility is also inconsistent within UK law. The MACR in England, Wales and Northern Ireland contradicts other status-based rights and responsibilities, such as the age of sexual consent, the right to buy tobacco, alcohol or to gamble, and political enfranchisement. It is in contrast with child welfare policies too, and diverges sharply from the response of the state when children aged under 18 are victims of offences, particularly familial violence or abuse. The Scottish Children’s Hearings System already adopts a more welfare-based approach to youth justice, and this results in injustice for children residing in each of the devolved nations of the UK.

Evidence does not suggest that contact with the criminal justice system (CJS) is an effective way to reduce crime. In fact, children who come into contact with the CJS at an earlier age are at increased likelihood of future offending [10]. It is difficult to disentangle the impact of risk factors, such as early disadvantage, family and parenting dynamics, and/or psychosocial characteristics [11] from the impact of involvement in the YJS itself [12, 13]. Many children who have experienced Adverse Childhood Experiences (ACE) do not become involved in offending or antisocial behaviour, whereas nearly 40 per cent of those who become involved in the YJS go on to re-offend [14, 15]. This suggests that the CJS itself is criminogenic; exposure to criminalised contexts increases the risk of re-offending.

In the 1990s and 2000s changes to law and practice drew increasing numbers of children into contact with the Youth Justice System (YJS). Since 2008 a positive youth justice approach has successfully been used, and reductions in children entering the YJS have been seen in England and Wales.

Find out more

Debates about the MACR are situated within wider social and political debates about the nature of crime and responses to harms caused by young people. In the UK the MACR has been strongly influenced by public fear of rare but serious offences committed by younger children in the 1990s [16]. The political response to public concerns included the abolition of doli incapax (the assumption that a child is incapable of forming criminal intent) in the Crime and Disorder Act 1998. Doli incapax had provided an element of discretion and protection when younger children were accused of committing offences. The 1990s also saw the introduction of legal measures against disruptive behaviours, such as Anti-Social Behaviour Orders, which disproportionately affected children [17]. Further, the Youth Justice Board for England and Wales (YJB) implemented a wide-reaching programme of ‘early intervention and prevention programmes’ for younger children considered at risk of involvement in offending behaviour. While these interventions led to positive outcomes for many children, they also had the unintended consequence of drawing more children into the YJS [18]. Similarly, the ‘Offences brought to justice’ performance-related target established for the police in 2002 led to a disproportionate focus on children and young people, whose offences are often more visible and less serious than those committed by adults [19]. However, since 2008, there has been a significant decrease in the numbers of first-time entrants (FTE) to the YJS and a reduction in the numbers of children in custody, suggesting a shift in this cycle [15, 19, 20]. In recent years, the YJB has adopted a positive, Child First approach, moving from a risk and responsibility model to a child-centred approach which responds to need [21]. In this context, a reconsideration of the MACR is timely.

2.2 Offences committed by younger children

Fewer than 10 per cent of children within the Youth or Criminal Justice system are aged under 14 [14]. These youth crime statistics capture cautions and sentences providing a record of police actions, so may not accurately represent child behaviour.

The full extent of the involvement of younger children in the youth justice system is difficult to assess. The annual YJB statistics provide snapshot data on younger children’s involvement in offending behaviour where it has resulted in a caution or a sentence in England and Wales. Children under 10 are excluded from this dataset exactly because they cannot receive a criminal outcome. In the year ending March 2020*, about 10 per cent of children (aged 10-17) who received a caution or sentence were aged below 14 (Table 1). The total number of 10-13 year olds in this group was 2,439, which is about 1 in every 1,200 children in this age group at the time [22]. Reflecting patterns across the YJS, the majority of these children were boys, and children from Black or mixed ethnic backgrounds were disproportionately represented (Table 2). These figures certainly underestimate the number of contacts between children and the YJS, as they record policing actions rather than child offending itself. The many changes in legislation and sentencing guidance, the impact of early intervention and prevention programmes, and different data recording practices across police forces and Youth Offending Teams (YOT) result in reporting differences.

*Although updated data from 2021 is available, this report focuses on data from 2020, published in 2021; the impact of the COVID-19 pandemic means that data from the year 2020-2021 is likely provide an atypical account of children’s involvement in the youth justice system

Table 1: the number of children aged 10-13 who received a caution or sentence in the year ending March 2020 by age and sex
Age (years) Girls Boys Total
10 4 16 20
11 12 113 125
12 100 407 507
13 254 965 1,220*
All aged (10-17 years) 2,829 16,188 19,026

* 1 unknown, Figures taken from Supplementary Table 3.6, YJB/MoJ Statistics 2021[14]


Table 2: the number of children aged 10-13 who received a caution or sentence in the year ending March 2020 by age and ethnicity
Ethnicity 10 years 11 years 12 years 13 years Total
Asian 2 4 30 52 88
Black 0 6 40 112 158
Mixed 1 7 57 111 176
Other 1 2 7 27 37
Unknown 2 0 12 25 39
Total BAME 4 19 134 302 459
White 14 106 361 893 1374

Figures taken from Supplementary Table 3.5, YJB/MoJ Statistics 2021 [14]

Offences committed by younger children tend to be less serious and result in fewer custodial sentences [14]. In the year to 2020 there were only 29 crimes committed by children aged 10-14 years in the whole of England and Wales judged to be in the most serious category.

Most proven offences committed by children aged 10-13 are relatively low-level, for example criminal damage, public order and theft (Table 3). A significant proportion are involved in violence against the person offences – although this category includes a wide range of offences, from verbal threats through to physical harm and many of the offences committed by younger children may be arguably less serious [14]. Gravity scores are a better estimate of seriousness of offence, which categorises proven offences on a scale ranging from least (1) to most (8) serious offences. Data on the gravity of proven offences committed by children is provided in aggregate age bands (Table 4). Younger children (aged 10-14) are less likely to be involved in higher gravity offences, with 85 per cent of offences committed by younger children having a gravity score of 1-3. In the year ending March 2020, only 29 offences committed by children aged 10-14 had a gravity score of 8 [14]. The number of individual younger children committing the most serious crimes (held under section 90, 91, 226 or 226B of the Powers of Criminal Courts (Sentencing) Act 2000) is not reported separately, but we can assume it is very rare. In contrast, 2.6 per cent of all children aged 10-15 years were a victim of a violent crime in the year to March 2020 [95].

Table 3: proven offences by children by offence type and age, year end March 2020
YJB offence type Age 10 Age 11 Age 12 Age 13
Arson 0 3 25 34
Breach of bail 0 0 1 19
Breach of Conditional Discharge 0 0 0 9
Breach of Statutory Order 0 0 14 44
Criminal damage 8 70 179 432
Death or injury by dangerous driving 0 0 0 0
Domestic burglary 1 2 11 42
Drugs 0 0 22 80
Fraud and forgery 1 1 3 12
Motoring offences  0 2 11 76
Non domestic burglary 1 2 19 47
Other 1 9 19 70
Public order 8 43 111 261
Racially aggravated 1 5 31 50
Robbery 0 4 21 122
Sexual offences 0 6 12 46
Theft and handling stolen goods 6 26 91 317
Vehicle theft/ unauthorised taking 0 1 26 80
Violence against the person 20 115 579 1,260
Total 47 289 1,175 3,001

Figures calculated from Supplementary Table 4.3, YJB/MoJ Statistics 2021 [14]

Table 4: proven offences by children by age and gravity score, year ending March 2020
  Proportion of offences committed by children aged 10-14 years Proportion offences committed by children aged 15-17 years
Gravity score 1-3 85.1% 79.0%
Gravity score 4-8 14.8% 20.9%
Total number of offences 11,101 37,969

Figures taken from Supplementary Table 4.5, YJB/MoJ Statistics 2021 [14]

2.3 The context for offences committed by younger children

Many of the children who commit offences at a young age have experienced considerable disadvantage [15], which is often compounded by their involvement in the criminal justice system. A growing understanding of the relationship between early adversity and trauma, and the role of sexual and criminal exploitation in particular, raises questions about whether a criminal response is appropriate for children who are or have been abused [23, 24].

Read more on this section

Limited information is available within official statistics or reports about the context of specific offences; this level of detail is either not collected or is aggregated, often without identifying inter-relationships between offence type, alleged perpetrator and victim/s. However, as noted above, research does provide evidence of a correlation between children’s prior experiences of discrimination, disadvantage, abuse and harm, and an increased likelihood of being involved in offending behaviour. Analysis of AssetPlus assessments, conducted by YOT professionals when a child becomes involved in the YJS, shows that a greater proportion of younger children in contact with their service have additional needs (speech, language and communication difficulties, mental health needs, learning and education) and parenting concerns, than among those aged 15-17 [15]. Combined with evidence that Adverse Childhood Experiences (ACE) and experience of trauma are triggers for poor mental health and disruptive behaviour indicates that the children with the most complex disadvantage may be particularly vulnerable to involvement in offending behaviour. Furthermore, the over-representation of specific groups of children (such as minoritised ethnic groups and/or care-experienced children) may reflect structural systemic factors that contribute to their involvement in the YJS [23, 25]. There is also growing evidence that children’s involvement in offending can result from child sexual exploitation (CSE) and child criminal exploitation (CCE), including being groomed and recruited into ‘county lines’ drugs offences, raising questions about whether such behaviour should be seen as offending or abuse [24].

2.4 The sentences received by younger children

Only small numbers of younger children are held within the youth secure estate [14], but they are involved in a higher number of negative situations within custody. Children aged 10-14 years have higher rates of Restrictive Physical Interventions, are more likely to be involved in assaults/fights and have higher rates of self-harm than older young people [14]. This suggests that younger children are less well-equipped to cope in the custodial environment.

Read more on this section

The YJB statistics on sentencing are not disaggregated by age, so it is difficult to know which sentences are most frequently used for younger children. However, given the relative level of gravity and type of offence committed by younger children, it is most likely that they will be sentenced to a Referral Order, with smaller numbers being sentenced to a Youth Rehabilitation Order. Very few children aged 10-13 receive custodial sentences, such as Detention and Training Orders or sentences under s90 or 91 of the Powers of Criminal Courts (Sentencing) Act 2000 [14, 15]. Indeed, the average monthly number of children aged 10-14 held in Secure Training Centres, in the year ending March 2020, was just 29 [14]. Department for Education data for 2021 [26] shows that 96 per cent of residents of Secure Children’s Homes were aged between 14 and 17, with just six children being aged either under 13 or over 18 years. In many years, the counts of children aged under 14 years in Secure Children’s Homes are very low (between 1 and 5 children), so the data is not listed.

Despite the small numbers of younger children held within the youth secure estate, they are involved in a higher number of negative situations within custody than their older peers. For example, children aged under 14 have higher rates of behaviour management interventions (Restrictive Physical Interventions, single separation, use of force), are more likely to be involved in assaults/fights, and have higher rates of self-harm than older young people [14, 15]. In the year ending March 2020, 62 per cent of children aged 10-14 years received Restrictive Physical Interventions (RPI) while in custody, receiving an average of 3.3 RPI compared to 36 per cent of 15-18 year olds who received 1.9 RPIs (Supplementary Table 8.5 [14]). Similarly, each month there were an average 51.3 incidents of self-harm per 100 10-14 year olds in custody compared to 21.4 incidents per 100 15-18 year olds (Supplementary Table 8.10 [14]). This suggests that younger children are less well-equipped to manage the challenges posed by the custodial environment and that their experience of custody is potentially more problematic and more harmful than that of older children.

2.5 Outcomes for younger children convicted of crimes

Contact with the youth justice system can lead to negative cycles of criminality. A criminal label can lead to stigma, discrimination and differential treatment, including by schools, the police and the judiciary, increasing the likelihood of offending again [20]. Those who are incarcerated experience disruption to education, loss of positive relationships, increased exposure to offending communities and drug use, together with a high risk of victimisation and abuse [27].

Find out more on the outcomes for younger children

Children involved in the YJS may experience labelling and stigmatisation, disruption to their family and peer relationships and have fragmented access to education, which may be compounded by delays in the justice process itself, all of which may contribute to poor child outcomes. There is concerning evidence of disproportionality and discrimination with the YJS [20].

For the minority of younger children who are held in secure custody, the impact of potentially harmful experiences within custodial institutions may further exacerbate the problems and difficulties faced. A review of literature suggests harmful outcomes for incarceration of children under the age of 18 [27]. Although evidence reported seldom examines differential impact by age, considering the nature of these impacts it is reasonable to suppose these would be greater at younger ages. Incarceration has a negative social impact through decreased contact with prosocial peers, acculturation into offender communities and the disruption of family and community contacts and networks. Health is directly harmed by poor conditions within custodial institutions (e.g. overcrowding, nutritional quality of food, exposure to violence, lack of opportunities for exercise and education). The existence of substance misuse, violence and sexual activity and sexual abuse within the custodial estate presents a risk to children. Victimisation and abuse are common experiences in custody, with some evidence that younger prisoners, those with a history of abuse or assault, those from minority ethnic groups, and those with a sexual orientation other than heterosexual are at greater risk. There are high rates of mental health disorders among children in custody, and the outcomes for these are worse following incarceration including in high rates of suicide and self-harm. The impact of the stigma of having been in custody, along with the impact of disrupted education and disrupted relationships has a long-lasting impact into adulthood [27-31].

2.6 Raising the Minimum Age of Criminal Responsibility (MACR) in England

There is a compelling case for raising the age of criminal responsibility. But this opens the question of what alternative approaches are appropriate and effective for children who cause harm to others.

This review of literature considers the evidence to support alternatives to criminal procedures when younger children (aged 8-13 years) act in criminal ways. A structured review approach was used to ask three questions:

  1. What works to reduce future offending and criminal convictions for children aged 8-13 years at high risk of involvement with the criminal justice system?
  2. What works to improve outcomes (mental health, educational, social care involvement) for children aged 8-13 years at high risk of involvement with the criminal justice system?
  3. What works to support victims of crimes committed by children under the age of criminal responsibility?
Read more of this section

Given the poor outcomes for younger children involved with the criminal justice system, the low level of crimes that result in convictions, the small number of children involved, and the appeal to both rights and reason that children aged under 14 are simply too young to be treated in the same way as older young people, there is a compelling case for raising the age of criminal responsibility. But this opens the question of what alternative approaches are appropriate and effective for children who cause harm to others.

Increasing the MACR recognises that a criminal response is not appropriate, but does not imply no action should be taken when a child does harm to others. It is already the case that many acts that could be criminalised are dealt with outside of the criminal justice system. Child safety orders, local child curfews and parenting orders all exist as alternatives to criminal proceedings. Children may be referred through a social care route to early intervention and prevention initiatives such as family hubs, parenting support programmes, or Youth Inclusion and Support Panels with the aim of improving the circumstances that have resulted in problem behaviours. We do not know how often any of these actions are currently taken in England; as noted by Phil Bowen, Director of the Centre for Justice Innovation, to the House of Commons Justice Select Committee on 30 June 2020 “The Youth Justice Board does no national data reporting on diversion cases, so at the moment we do not know enough about who the kids are and what their outcomes are.” [32] There is good reason to believe that large numbers of families receive these kinds of responses. For example, over 115,000 families received support through the Troubled Families programme between 2012 and 2015 and many more have been supported since, through this and the Supporting Families Against Youth Crime Fund that was launched in 2019. The latter funded a range of interventions in 21 local areas, including school-based activities, diversionary schemes, mentoring and whole family therapy [33].

Having raised the MACR in 2021, guidance in Scotland [34] emphasises that the need for effective investigation of harmful behaviour by children is still essential, but that all interactions with children should be based on a trauma-informed approach that recognises the child’s wellbeing as a primary consideration. The guidance stipulates that, where appropriate, child protection measures or ‘Early and Effective Intervention’ processes should be instigated [34]; because the child is below the MACR, there is no criminal justice system involvement, thereby avoiding the risks of potential criminalisation.

This review of literature considers the evidence to support alternatives to criminal procedures when younger children (aged 8-13 years) act in criminal ways. A structured and systematised review approach* was used to ask three questions:

  • What works to reduce future offending and criminal convictions for children aged 8-13 years at high risk of involvement with the criminal justice system?
  • What works to improve outcomes (mental health, educational, social care involvement) for children aged 8-13 years at high risk of involvement with the criminal justice system?
  • What works to support victims of crimes committed by children under the age of criminal responsibility?

 

Evidence about what works for younger children in contact with the criminal justice system was drawn from systematic reviews of effectiveness studies (section 3.1), from a review of practice literature in the UK (Section 3.2) and from Case Studies of current practice (section 4).


*For further details of the review approach contact the first author Patricia@PJLulcas.org

3. What works for children under the age of 14 in contact with the criminal justice system?

This review draws on two sources of information: evidence from Systematic Reviews of international research evidence, and evidence from UK practice.

3.1 Evidence from Systematic Reviews

3.1.1 Introduction

Some of what we know comes from research with older children. A wealth of research with older children suggests that exposure to prison facilities (scared straight), processing through the juvenile justice system and custodial sentences all increase the likelihood of future offending [35, 36]. Given these more punitive approaches have been shown to be harmful in older children, they are likely to be harmful to younger children too, and we do not recommend them.

Similarly, whole-system approaches (for example, media campaigns or improvements to the built environment) do not target younger children in particular. However, systematic reviews of this evidence conclude that there is not enough evidence to judge whether or not these approaches are effective [37, 38].

Read more

There is a wealth of research published concerning how to respond to behaviour difficulties in younger children, or how to respond to antisocial behaviour in older children. For example, the College of Policing crime reduction toolkit divides evidence between ‘children’ or ‘young people’. There is little that addresses those aged 8-13 years, and we focussed our efforts on locating research specific to this age group. We prioritised systematic reviews of international research evidence (see section 3.1), and evidence from recent practice in the United Kingdom (see section 3.2).

Our search for reviews that allow examination of what works for children aged 8-13 years means that some of the most famous examples of evidence for what does not work are not included. Both organised visits to prison facilities (scared straight) [35] and processing of cases through the juvenile justice system (rather than through a non-justice route) [35] have been found to increase rates of future re-offending. Similarly, a majority of studies show that re-offending rates are lower for children who receive non-custodial vs custodial sentences (although this has not been established in controlled studies) [36]. These reviews do not allow us to differentiate findings for studies of children in this younger age group, but it is plausible that their findings will also apply here. Given these more punitive approaches have been shown to be harmful in older children, they are likely to be harmful to younger children too and we do not recommend them.

Our focus on direct work with children also means that whole system approaches may be underrepresented in the research we located. For example, systematic reviews of the impact of improvements to the built urban environment [38] and media campaigns [37] on youth violence are not included here. However, neither review found sufficient evidence to draw firm conclusions on their effectiveness [38] [37]. Another systematic review suggests that formalised violence prevention collaborations (e.g. between community organisations, local or national government, schools and universities) play a vital role in stimulating system level policy change [45], but did not report on outcomes for children themselves.

We located 15 reviews published between 2010 and 2021 which examined programmes and interventions for children at high risk of involvement in the CJS and allowed us to consider outcomes for children aged 8-13 years specifically. Few reviews compared outcomes between age groups of children.

 

We located 15 reviews published between 2010 and 2021 which examined programmes and interventions for children at high risk of involvement in the CJS and allowed us to consider outcomes for children aged 8-13 years specifically. These reviews considered youth offending, criminal recidivism, aggressions, Conduct Disorder (CD), delinquency (broadly defined), behaviour problems, school exclusion and child mental health outcomes. These reviews are summarised in Table 5. Most reviewed psychosocial approaches including cognitive and behavioural programmes, systemic family therapy, treatment foster care and mentoring and it is reasonable to suppose there is considerable overlap between these reviews. The effectiveness of police-led diversion, curfews, drug treatment courts, restorative justice and pharmacological treatments were also reviewed. The reviews overlap in included studies, often include a mix of approaches, and child outcomes. The Appendix provides additional methodological details about included studies (available on request).

Across the literature we found that most reviews include a wide range of child participants, and that very few compare outcomes between age groups. In most of the reviews included here, it was possible to identify the ages included in each contributing study so that child age could be discussed even where review authors themselves did not comment on this.

The evidence presented does not separate out what works to improve outcomes for younger children who are accused of the most serious crimes. This is probably due to the very small number of children under 14 to whom this applies. High quality research-based evidence relies on rigorous evaluation with large numbers of cases, which is not possible for rare events and, fortunately, the number of serious harms caused by younger children is extremely small. We also found no review of practice with those children convicted of the most serious crimes. While caution should be used extrapolating from rare events, it would be helpful to collect and report more data about both victims and child offender outcomes in the case of the most serious offences.

3.1.2 Psycho-social approaches

Psycho-social approaches address the relationship between social factors (e.g. parenting, relationships, social functioning and social context) and a child’s individual thoughts and behaviours.

Evidence from multiple systematic reviews shows that family systems approaches (such as family therapy and Multi Systemic Therapy) are likely to be very effective at reducing disruptive behaviour, antisocial behaviours, Conduct Disorders (CDs) and substance use. There is some evidence that they will be more effective in younger children [46-49].

Read more about psycho-social approaches

Family therapy, individual therapy and family systems approaches were reviewed in a number of studies, most of which reviewed a number of different approaches.

Granski and colleagues [46] undertook a meta-analysis to examine gender differences for programmes aiming to reduce disruptive behaviour problems (broadly defined). Studies included cognitive skills training, behaviour modification and family systems approaches. Additionally, they compared outcomes for children aged 11 and younger to those aged 12-18 years. While this age cut-off divides the age group of interest in this review, it provides important evidence where few meta-analyses are available. They found that cognitive systems or family system interventions were more effective than other approaches for both girls and boys. In addition, programmes that were longer, introduced at a younger age, and that worked with both children and their parents were more effective. However, programmes varied considerably, and that positive impact waned over time. Nonetheless, this suggests that family systems approaches in younger children are very likely to be effective at reducing disruptive behaviour.

Sawyer and colleagues examined the impact of a range of individual and family-based interventions on long-term antisocial behaviour (up to 35 years post intervention, but 4 years on average) [47]. This review included both very early preventive work (e.g. home visiting for infants) and later treatment programmes (e.g. functional family therapy within the justice system). In their meta-analysis authors found a modest reduction in antisocial behaviour across all approaches. This size of the effect increased when only official records (ie criminalised behaviours) were included. Child age itself did not significantly change the impact, but it did interact with other programme elements. Programmes that included a peer-group element were less beneficial for preadolescents and adolescents, although beneficial when used with very young children. There was some evidence that therapeutic interventions (for those with diagnosed or established problems) had greater impact than preventive programmes, but differences in study population, context and type means it is not clear what causes this difference. Individual studies where the average age of participants was 8-13 years show similar mixed findings, suggestive of an overall modest positive effect.

Bakker and colleagues [50] found a positive effect of psychosocial (eg psychotherapeutic and social cognitive) treatments on symptoms of CD. The impact on children less than 10 years was larger than among children older than 10 years, although this was based on parent-rated outcomes. Just 3 of the studies included in this review included participants aged 8-13 years. However, although reductions in most conduct symptoms were reported, in most cases the impact was very small.

Von Sydow and colleagues [48] looked at 42 studies testing systemic therapy on CD and delinquency in controlled trials. The authors do not distinguish outcomes by age group. The authors conclude that these studies provide strong evidence that systemic therapies have been effective in reducing CDs and substance use across studies. Many of the included studies include children aged 8-13 years, although few were specific to this group, and they also find reductions in delinquency and substance use.

These findings are confirmed by Waddell and colleagues [49] who note positive improvements in child behaviour outcomes following Multi-Systemic Therapy and parent-interaction therapy used as a treatment for children with behavioural problems. However, while all of the studies included children in the range 8-13 years none were specific to this age group so we cannot say how child age may have modified this effect.

Behavioural approaches (e.g. behavioural modelling, behavioural contracts) have been shown to be effective for reducing persistent offending behaviour for all ages of children where this has been tested. Programmes carried out in court settings were less effective than community programmes [51].

De Vries and colleagues reviewed a wide range of approaches to try to identify the “effective ingredients” of programmes to reduce persistent offending behaviour [51]. Their review included a number of different psychosocial and behavioural interventions, treatment foster care, mentoring and restorative justice. They suggest that behavioural modelling, parent skills training and behavioural contracting resulted in larger reductions in re-offending than other approaches although the impact was still modest. They also found that these programmes appear to be equally effective in younger age groups. However, programmes carried out in court settings were less effective than those in community or home settings. 

One systematic review of online parenting programmes for parents of children with diagnosed behaviour problems found modest benefits, but most of the studies reviewed were with children younger than 8 years old [52].

There is an established and large evidence base for parenting programmes as a preventive programme when used with parents of very young children with behavioural difficulties [49, 53]. Few programmes are specific to the age group of interest here. One systematic review did include studies with this age group, using online parenting programmes for children and adolescents with diagnosed behaviour problems [52]. Authors conclude that online parenting interventions were effective in reducing diagnosed behaviour problems. However, the impact was modest and most of the evidence comes from studies including only younger children.

A systemic review of Cognitive Behavioural Therapy (CBT) reports a large reduction in symptoms related to Conduct Disorder, Oppositional Defiant Disorder and Attention Deficit Hyperactivity Disorder [54]. However, the studies included in this review were less rigorous than in other reviews so the evidence is less convincing.

In a review of Cognitive Behavioural Therapy (CBT) for CD and Oppositional Defiant Disorder (ODD) symptoms, Riise and colleagues report a large reduction in symptoms of both ADHD and CD/ODD [54]. At post treatment, remission rates were 38 per cent for ADHD and 48 per cent for CD/ODD. Authors also conclude that more severe symptoms at pre-treatment were associated with a larger effect size. This review is important because many included studies had a range or mean age within the age group of interest, so the evidence for the 8-13 year old group is comprehensive. However, the study types included in this review are broad, and where there is no comparison group and outcomes are teacher or parent reported there is a strong chance of responder bias in these studies so the evidence is less convincing.

Mentoring may have positive impacts on aggression, drug use, delinquency and academic functioning. However, the authors of the systematic review also note that mentoring models vary widely between studies, and so does effectiveness [55]. So while there is evidence mentoring may be helpful in this age group, it is not clear which form of mentoring may be helpful.

Tolan and colleagues [55] examined the evidence for mentoring to reduce delinquency (for example number of chargeable offenses) or aggression, drug use and academic functioning. Around half of the included studies studied mentoring in the 8-13 year old age group. Authors conclude that mentoring has a modest, positive impact on these outcomes. While authors do not compare effectiveness by age of participating children, the studies which include children aged 8-13 years old also tend towards positive results although not all do so. Although this is a thorough and well conducted review, the studies themselves vary considerably in their approach to mentoring and in their research methods. In particular, included studies seldom described the mentoring programme in detail. Considering the large variations in study outcome, this means it is not possible to know which form of mentoring is likely to be effective.

A range of school-based interventions (academic skills, counselling, mentoring/monitoring, and skills training for teachers) were shown to be effective in reducing school exclusions in the short term, but not in reducing antisocial behaviour [56]. The impact on exclusion rates was modest and not sustained.

Valdebenito examined school-based intervention to reduce exclusions from mainstream schools [56]. Exclusion from school is a risk factor for later criminal involvement, so it is an important outcome to consider. Nearly half of the participants in studies they include were middle-school children, so findings are robust for the age group of interest here. They found that enhancement of academic skills, counselling, mentoring/monitoring, and skills training for teachers all reduced rates of exclusion but not antisocial behaviour. The reduction in exclusion rates was not sustained, and was modest in size with some evidence of greater impact on in-school rather than out-of-school exclusions [56].

3.1.3 Criminal Justice Approaches

Police-led diversions schemes, using cautions, restorative actions, and referrals to other services to avoid arrest, have been shown to reduce re-offending rates by about 20 per cent relative to usual practice [57].

Wilson and colleagues [57] reviewed police-led diversion approaches. Police-led diversion in these studies includes issuing cautions in the place of arrests, often accompanied by referrals to other services and/or police-led restorative actions. This approach was tested in 19 evaluations, only three of which certainly excluded children under the age of 14 years. In some studies more serious crimes were not included in the trial, but progressed to arrest as usual. Police-led diversion is generally positive for low-risk young people – with lower re-offending rate than those who experience ‘traditional’ processing [57]. On average across the studies, the re-offending rate amongst young people who received a caution was about 20 per cent lower than in the comparison arm (that is, about 8 re-arrests in the diversion group for every 10 re-arrests in the comparison group). The authors do not differentiate outcomes by age of child involved, so although we know that several studies included younger children it is not clear whether they experienced this scheme differently.

Youth curfews, where young people are not allowed in public spaces during night-time hours, do not appear to be effective in reducing children’s involvement as either victims or perpetrators of crime [58].

Wilson and colleagues reviewed [58] the impact of youth curfews, where children below a certain age are not allowed in public places during night-time hours, on both rates of crime by children, and on victimisation of children. They found that curfews have no effect on the overall number of crimes committed by children or the number of children who are victims of crime. In fact, during the hours of the curfew, a small increase in crimes committed children is seen (which may be explained by those breaking the curfew being stopped by police).

There is some evidence to suggest Drug Treatment Courts improve outcomes for substance using youth. A systematic review found a modest reduction in repeat offending among those who completed the programme [59], but this model is not used in the UK.

The effectiveness of drug treatment courts was assessed in one systematic review [59]. Juvenile Drug Treatment Courts (DTC) are largely a US phenomenon, although they have been used in Australia and New Zealand. They address adolescent substance abusers who also engage in criminal behaviours. They are specialised courts that can require a one-year drug treatment approach which may include psychosocial interventions, educational classes, mandated school attendance rates, urine tests, but will always include regular court appearances to review progress. Although they largely focus on the young person, they often mandate involvement of parents or carers. The authors of this review find that mean recidivism rate for those who successfully completed the programme was 16 per cent per cent compared to 21 per cent in non-completers. They further find that, although age was not related to recidivism rates, younger age was correlated with higher graduation rates. However, it is not possible to know from this study what ages of young people were included in the primary studies and therefore what ‘younger age’ means. The search conducted was also narrow, and it is unclear if any studies from outside the USA were included, and we therefore conclude that this evidence is weak for our purposes.

While there is a strong evidence base to suggest restorative justice approaches are effective at both reducing repeat offending and improving victim satisfaction rates [60], the effectiveness when used with 8-13 year olds in particular has not been included in a systematic review to date.

Restorative justice is included in a broad range of approaches in the review by de Vries and colleagues [51]. However, they do not draw out any specific findings about restorative justice. While there is a strong evidence base to suggest restorative justice approaches are effective at both reducing repeat offending and improving victim satisfaction rates [60], the effectiveness when used with 8-13 year olds in particular has not been reviewed and established.

3.1.4 Pharmalogical treatments

There is some evidence that in the short-term use of risperidone may reduce aggression and conduct problems in children and youths with disruptive behaviour disorders and that clonidine and guanfacine have small effects on oppositional behaviour in children with ADHD. However, two systemic reviews also find high rates of side effects including weight gain, fatigue and sedation [49, 61].

Read more about pharmalogical treatments

The impact of pharmacological approaches has been examined in four reviews, examining the impact of antipsychotics, mood stabilisers, and psychostimulants for aggression and symptoms of CD [49,61-63]. The studies included in these reviews largely overlap, and although all studies included cover the age range of interest, all but one are for a wider age range. There is some evidence that in the short term risperidone may reduce aggression and conduct problems in children and youths with disruptive behaviour disorders [61] including those with ADHD [62]. There is some evidence that clonidine and guanfacine have small effects on oppositional behaviour in children with ADHD [63]. However, there is also evidence that risperidone is associated with significant weight gain [61] and Waddell suggests that there are high rates of side effects in most studies including fatigue and sedation associated with risperidone [49].

3.2 Evidence from UK practice

While many of the interventions and approaches described in the UK grey literature could be applied to children aged 8 – 13, few targeted this age group. We also found few programmes had been subject to a rigorous evaluation.

As with the research evidence, when we searched the UK grey literature for examples from recent practice we found that while there were many interventions and approaches implemented in practice that could be applied to children aged 8 – 13, few were either specifically or certainly used for this age group. Often children aged 8-13 were either included at the older end of a younger age group (for example, programmes aimed at 7-11 year olds), or the younger end of an older group (for example, 12-17 year olds) or within a much larger cohort of all children aged under 18. We report here on those programmes which we are confident include children in the age group of interest, excluding those where it was simply not clear. We were comprehensive in our search for programmes, but we were not exhaustive. Many of the interventions we identified are no longer active, usually due to changes in funding or local priorities. Few have been subject to rigorous evaluations, but we highlight any evaluative evidence we found, and the strength of evidence rated using Early Intervention Foundation (EIF) standards [64].

3.2.1 Psycho-social approaches

Action for Children’s Serious Organised Crime Early Intervention Service works with ‘perennial non-engagers’, to provide intensive case work, whole-family work, and work with peers. An evaluation is ongoing [65].

Read the approach and evidence

Approach: Although this intervention covers a wider age range (11-18 years), we know they worked with children as young as 13 years so it is included here. The service has been operational in Glasgow since 2013 and in four other sites since 2019. Aimed at those known to statutory services as ‘perennial non-engagers’ and where previous attempts at diversion away from criminality have been unsuccessful, the service adopts a holistic approach in working with young people who will often have complex needs and high levels of risk. 

The project offers a bespoke service, identifying the individual drivers and causes of young people’s offending with the aim of empowering young people to make positive change. This is supported by three core elements of service delivery. First, provision of intensive case work such as 1:1 support, peer mentoring and evidence informed approaches. This is supported by the development of multi-agency working aimed at system-level change through information sharing, capacity building and developing effective responses to young people on the edge or involved in serious crime. Second, community-based early intervention and prevention with the young person’s wider network to deter and/or divert siblings, peers and associates away from criminality. Third, the project offers a whole-family approach working with family members and empowering them to support the young person’s journey onto more positive pathways.

Evidence: Not yet reported, a two-year evaluation began in February 2021 [65].

In Northern Ireland, the Early Intervention for the Prevention of Offending projects for 8–13-year-olds tested individual case work delivered by specialised staff delivered over 6-12 weeks [4, 66]. There is weak evidence of a reduction in the likelihood of children being known to police.

Read the approach and evidence

Approach: The Health and Social Care Board in Northern Ireland initiated 5 related projects aimed at preventing crime and antisocial behaviour. They provided funds for three agencies (NIACRO, Extern and Action for Children) to run five projects building on their existing programmes of work and ways of working. The common theme was individual case work over a relatively short period of time (6-12 weeks) delivered by highly trained staff delivering individual intervention work.

Evidence: EIF Level 2. Evidence of impact is drawn from monitoring records. In total 409 children were referred through this service in the two years of the evaluation period. An analysis of completed cases only suggests successful outcomes. While 50 per cent of children were known to the police at admission to the programme, only 7 per cent were known to the police at discharge. Those who worked with the children judged that their risk profile improved considerably during the intervention period. Report authors suggest positive improvements in behaviour, school attendance and in parental capacity to deal with behaviour. Caution should be used when interpreting these findings given the absence of comparison data, and the limited use of objective methods of data collection. The cost of the programme was estimated at £4,610 per child [4, 66].

Preventing Youth Offending Project (PYOP) worked with children aged 7 years upward at high risk of offending behaviour using a combination of individual and group work, together with art, music and drama workshops. There is preliminary evidence that offending behaviour decreased relative to a comparison group [67, 68].

Read the approach and evidence

Approach: This was a highly responsive programme for children aged 7 or older involved in, or at risk of involvement in, offending behaviour. Referrals could be from professionals or from parents or children themselves, if the child was identified as a prolific offender, was facing a custodial sentence, had particular offending needs (for example, being accused of sexual offences), or were considered to be at risk of involvement in offending behaviour. The programme was specifically tailored to the needs of the individual but could include:

  • Individual interventions, such as mentoring for reintegration into education, anger management, constructive use of time
  • Group work focusing on antisocial behaviour, problem solving, anger management, victim awareness, interpersonal skills, substance misuse, appropriate sexual behaviour, health issues
  • Workshops (art, music, drama) and outdoor activities to develop social skills and self-esteem

Siblings were welcome to join some of the activities and parents could be referred to counselling and other services.

The length and intensity of interventions varied according to the needs of the child (rather than the risk they appeared to present) with between five and 390 hours of intervention during the first six months. Younger children received the greatest levels of input.

Evidence: EIF Evidence level 2 (evaluation using a comparison group). Sixty-seven children and young people, with an average age of 13 years, had been involved in the programme. This group included young children who had been involved in very serious offences. The participants’ risk of offending (measured using a standardised risk assessment tool) reduced; while the comparison group’s risk of offending increased. No one intervention type was more effective than another. There was a strong association between the amount of input and the reduction of risk [67]

[68]. Although encouraging, this is preliminary evidence which needs confirmation in a more robust evaluation.

Enhance Case Management (ECM) introduces a Trauma Recovery Model to YOT, increasing staff knowledge of attachment, trauma and ACEs together with multi-agency case work. An evaluation is ongoing, but there is no current evidence of impact [69].

Read the approach and evidence

Approach: Although this intervention covers a wider age range (10-17 years), we know that 4 of the 21 young people in the evaluation study were under 14 years old so it is included here. Enhanced case management is informed by the Trauma Recovery Model, and provides YOT practitioners with additional training into the relationship between attachment disorders, experience of trauma, and Adverse Childhood Experiences and later offending, aggressive or antisocial behaviour. Multi-agency case formulation and intervention planning is used, and interventions address the development and mental health needs of the young person.

Evidence: EIF Level 2. Use of routine data without a matched comparison group. A matched comparison group was generated of 14 young people matched for reoffending risk and 10 young people matched for risk of serious harm to 21 young people who entered the enhanced case management process. However, the outcomes for this matched group are not reported in the evaluation report. For most children in the programme, there was no change in their risk level. The ECM has mixed evidence for impact on offending, with 9 of 21 not reoffending, 7 committing an offence of the same or lesser seriousness, and 4 committing an offence of greater seriousness. It is not possible to judge whether ECM had a positive impact on offending on the basis of these limited data.

ECM is currently being trialled in four YOTs in South West England in a collaboration announced in 2018 between NHS England, Public Health England, the local clinical commissioning group and Exeter University [69].

The POWER (Promoting Opportunities with Emotional Resilience) Project is one of the very few programmes that targets 8-13 year olds at risk of criminalisation. It is too early to know whether this is an effective intervention.

Read the approach and evidence

Approach: The POWER (Promoting Opportunities with Emotional Resilience) Project supports children and young people aged 8-13 years who are considered to be at risk of criminalisation. See the Case Study for further details

Evidence: EIF Not Level 2. No formal evaluation has been undertaken, and as yet there is not sufficient monitoring data to report on outcomes for children.

Mentoring programmes are delivered in a variety of models and ages of children, but around half explicitly targeted ‘vulnerable’ young people. A review of 366 programmes used in England suggests modest positive improvement for social, emotional and educational outcomes [70].

Read the approach and evidence

Approach: Mentoring programmes vary considerably in their operation, but a review by the Children’s Commissioner found that across England over half of the programmes explicitly targeted ‘vulnerable’ young people, including children in care, ‘disadvantaged’ young people, and those not in education, training or employment, although none explicitly focused on those at risk of offending behaviour. Most of the mentoring projects worked with older children (over the age of 12 years), but 68 included primary school age children.

Evidence: Unclear, this evidence is drawn from a rapid review of 366 mentoring programmes so evidence level will vary between programmes. The literature suggests modest positive improvement for social, emotional, and educational outcomes. There was considerable variation in impact within and between schemes [70].

Youth Inclusion and Support Panels are multi-agency panels that deliver individually tailored interventions for children aged 8-13; there is some evidence of positive impact on prevention of offending, as reported by parents [71].

Read the approach and evidence

Approach: Youth Inclusion and Support Panels (YISPs) are multi-agency panels, established by the YJB in 2002, which aim to identify and support young people aged 8-13 who were at risk of involvement in offending or antisocial behaviour. YISPs develop an individually tailored package of interventions for children and their families, which usually include support to access mainstream services (such as CAMHS, education support, housing and/or parenting support) and one-to-one interventions with a key worker; involvement in the YISP is voluntary and typically lasts for 6-12 months.

Evidence: EIF Level 2. Use of routine data without comparison group. The evaluation of the pilot YISPs [71] identified variable practice in terms of panel membership and approach, with some focusing on preventing crime and antisocial behaviour but others adopting a more long-term, welfare-based approach. There was not always a clear link between the outcomes of assessments and the support offered to the child, with a tendency to focus on only those risks that could be addressed. Inconsistencies in assessments and data collection made a quantitative analysis difficult, but there was evidence of a reduction in risk, particularly for children with higher initial levels of risk, but it was not possible to identify whether any specific interventions were more effective than others. The role of the keyworker was a critical factor however, with regular, intensive and consistent support being highly valued. The evaluation highlighted that the majority of parents reported their child had not been involved in antisocial or offending behaviour during their involvement with the YISP and were optimistic that they would not offend in the future. It was noted that many of the families involved had previously asked for support but only received it once behaviour worsened; they may have benefitted from being referred to the YISP at an earlier stage.

3.2.2 Criminal Justice Approaches

A Diversion, Prevention and Youth Justice model working with police and YOTs to divert child cases away from criminal outcomes [72]. There is weak evidence from a before and after comparison that suggests a reduction in statutory referrals, a reduction in offending, and a reduction in First Time Entrants in the study area. Researchers estimate police avoided £146,741 in costs [73].

Read the approach and evidence

Approach: Although this approach covers a wider age range (8-18 years, mean age 14 years), more than 250 referrals were for children aged 9-13 years so it is included here. The Diversion Programme was introduced in 2016 and used an integrated decision-making process between Suffolk youth offending teams and seconded police officers. It aimed to avoid formal sanctions against minor delinquency amongst 10- 17-year olds, diminish the number of first time entrants (FTE) into the criminal justice system (CJS) and reduce reoffending rates. A preventative strand for children aged eight and above at risk of future offending was included. It is similar in approach to other diversionary schemes such as the Swansea Bureau, the Youth Justice Liaison and Diversion Scheme in Liverpool [74] and the Durham Model devised by the County Durham Youth Offending Service (CDYOS). Each child referred to the scheme is assessed, and according to their risk level is assigned to different intervention levels. Those judged to be a low risk of offending/re-offending, received short intervention (6 weeks, fortnightly), which may include a restorative element; those judged medium risk receive longer and more intense intervention (8 – 12 weeks, weekly) and those with highest risk receive 12 – 26 weeks of weekly intervention.

Evidence: EIF Level 2. Use of routine data without comparison group. Over the 2-year period, 710 young people went through the diversion process (some more than once). Researchers observed a reduction in statutory referrals, a reduction in offending among those referred to the scheme, and a reduction in First Time Entrants to the CJS. Researchers estimate the programme avoided £146,741 in police costs over its lifetime [73].

Information was collected from 115 YOTs in England who report using point-of-arrest diversion schemes [75]. The schemes varied but tended to include short assessments with arrested children and quick referrals into light-touch, voluntary interventions. Although there is a research base to suggest this approach can be successful in reducing crime, few schemes were monitoring re-offending data and this report did not include evaluation data.

Read the approach and evidence

Approach: The Centre for Justice Innovation mapped the extent of youth diversion projects across England and Wales. Of 152 YOTs contacted, 115 YOTs respond and confirmed that they operated a point-of-arrest diversion scheme. This indicates very widespread use across England and Wales. The schemes varied but tended to include short assessments with arrested children and quick referrals into light-touch, voluntary interventions. More than half of the schemes include victim awareness classes, and most (66 per cent) involve victims in the scheme.

Evidence: Unclear, this evidence is drawn from a survey of 115 YOTs. There is a strong international evidence base that demonstrates that youth diversion can reduce crime, is cost-effective, and creates better outcomes for children, although few of the schemes identified by CJI collected such data or were formally evaluated [75]. Only half of these answered questions about outcome and monitoring, of whom only 45 per cent confirmed that they collect data on reoffending rates. Authors suggest that there is low awareness of monitoring processes.

Youth Crime Reduction and Sport Pilot offered sports-based sessions to young people living in areas with high levels of antisocial behaviour [76]. One in five of those involved were aged 10-13 years. There was weak evidence of mixed success, with 3/5 areas experiencing a reduction of crime, one no change, and one an increase.

Read the approach and evidence

Approach: Although this intervention covers a wider age range (10 years and upwards with no upper age set), 20 per cent of those they worked with were aged 13 years or under so it is included here. The project ran for the 2 years to March 2017. The project tests sport-based interventions to reduce youth crime. The projects were diverse and each project reflected the local context in which it was developed. Typically, projects aimed to engage at least 20 participants in a sport-based project one evening per week (two projects offered two sessions per week) in an area that was identified as experiencing ASB issues. Football was the sport offered by most projects (n=5). Other sessions utilised boxing, dance and multi-sports.

Evidence: EIF Level 2. Use of routine data without comparison group. Statistically significant reductions in youth-related ASB were found in two areas, a non-significant reduction in one area, no change in one area, and a small increase in the fifth area, although without comparison areas or data no causality can be attributed. Findings are not disaggregated by age, so it is not clear if the youngest children had different outcomes. Qualitative findings suggest positive experiences for young people [76].

3.3 Summary of evidence

The strongest evidence for reductions in offending behaviour are for systemic therapeutic approaches such as family therapy and Multi Systemic Therapy. They are very likely to reduce disruptive behaviour, antisocial behaviours, CD and substance use in children aged 8-13 years. This conclusion is supported by rigorous systematic reviews of research evidence. These approaches are also successfully implemented in practice in the UK, including as part of individually tailored, trauma informed, responses to children.

Behavioural approaches (e.g. behavioural modelling, behavioural contracting) have also been found to reduce offending in research programmes. These are more effective when delivered in a family or multi-modal format.

Police-led diversion schemes, using cautions, restorative actions, and referrals to other services to avoid arrest, have been shown to reduce re-offending rates by about 20 per cent relative to usual practice. In many of the effective programmes actions are used in combination, for example police use cautions alongside referrals to other services or as part of a restorative approach. However, the evidence does not suggest whether the impact is the same or different in younger compared to older children.

While restorative justice approaches have been shown to be effective in reducing reoffending and improving victim satisfaction, the research evidence reviewed here did not examine whether impact on younger compared to older children. Given their developmental stage, we need more evidence to be confident this approach is also effective for children aged 8-13 years.

There is no evidence to suggest that curfews applied to all young people are effective in reducing either child offending or child victimisation. Evidence from older children suggests that responses from within the criminal justice system are less likely to be effective, and programmes which increase contact with prisons (both as part of ‘scared straight’ programmes or when comparing custodial to non-custodial sentences) increased offending.

Other outcomes for children considered in the research literature reviewed were largely closely related to offending. For example, school exclusions, problem behaviour, symptoms of Conduct Disorder, aggressive behaviours.  As with re-offending, there is evidence to suggest that a range of psycho-social approaches (approaches addressing the child’s psychological and/or social context) improve these. Systemic therapies, cognitive behavioural therapy, parenting programmes , skills-based programmes for teachers, and mentoring all have some evidence for success, however the impact on this particular age-group and context is not always commented on. The evidence for mentoring, in particular, is difficult to interpret, programmes vary considerably as does impact, and results for this age group are inconsistent.

The stronger evidence for more intensive, family-based work found in this review of evidence reflects what we know about the close relationship between adverse childhood experiences and disruptive behaviour.

4. What works to support victims of crimes committed by children under the age of criminal responsibility

Victims feel most supported when their voices are heard and the harm to them is taken seriously. We find no evidence that the age of alleged child perpetrators is considered in how we should support victims of crime.

It is essential that responses to crime consider the impact on victims and their families. There is little direct evidence to suggest we should respond differently to victims according to the age of the perpetrator.

Schools are common sites of harm, which means children are often both victim and perpetrator. School bullying programmes often have a restorative element, recognising that children will continue to be educated together and so the impact on the victim must be addressed.

Adult victims are often known to perpetrators, including parents, carers, and professionals working closely with them. The criminal justice system is not suited to tackling intra-familial conflict and the core issue is not the age of the child, but the limits of legal responses to the problem.

For victims, process matters as much as outcome There is little evidence that victims consistently demand stricter punishments. Restorative justice is valued by victims, particularly when they perceive offences are related to immaturity. However, younger children involved in harm or offending, may not have the emotional maturity or communication skills needed to fully engage in restorative conversations.

Find out more about victim support

Responses to crime must consider the impact on victims and their families. While rare, children are sometimes involved in serious or persistent crime that can have a significant impact on victims. The lack of data within the YJS differentiated by age of alleged offender means that we cannot say with certainty whether victims’ experience is different when younger children commit crimes. We found no research reports or examples from practice where support was provided differently to victims when perpetrators were younger children. As one practitioner reported “all victims have a consistent response, no matter the age of the child who is the alleged perpetrator”; some others acknowledged concerns about younger children’s ability to enter into restorative interventions.

There is little direct evidence to suggest responses to victims should differ according to the age of the perpetrator. We found no systematic reviews, evidence reviews, or recent practice literature examining the impact of crimes committed by younger children on victims. However, there are particular features of the context for troubled behaviours in younger children that may be important when considering the experiences of victims. The victims of crimes committed by children are often children themselves, and the locations of crimes are often school or home.

Schools are common sites of harm, which means children are often both victim and perpetrator. While adults might view some of the offences which happen in schools as minor, the levels of harm caused can be significant particularly where there is cumulative from persistent offending [78, 79]. Schools themselves often manage offences committed in school (or on the way to or from school) without the involvement of police or other criminal justice agencies, including some more serious offences [78]. Serious offences are sometimes prosecuted, particularly if the child or parent/guardian requests that it is recorded as a crime [78]. Most research to date focusses on the school environment as a contributory or protective factor for criminal behaviour [80]. Research considering the impact on victims finds that harms include feeling frightened, avoiding school, reduced engagement in extra-curricular activities and lower levels of achievement [77, 81]. School bullying programmes often have a restorative element, recognising that children will continue to be educated together and so the impact on the victim must be addressed.

Adult victims are often known to perpetrators, and those who commit more serious crimes are more likely to be known to the victim in all cases [82]. Adult victims of harms committed by children include parents, carers, and professionals working closely with them (such as teachers, health or social care staff). Although no data are available, physical, emotional or financial abuse committed by children on parents is reported relatively regularly by practitioners [83]. Social workers often suggest that parents involve the police, but the criminal justice system is not suited to tackling intra-familial conflict [84]. Legal remedies available through civil or family justice systems, such as injunctions or non-molestation orders are problematic, with all interventions potentially exacerbating relationship difficulties and include the risk of the child being removed from their home. The core issue is not the age of the child, but the limits of legal responses to the problem. All currently available options have significant negative consequences for the child, their parent/s and wider family, regardless of the age of the child.

For victims, process matters as much as outcome. There is good evidence to suggest that the availability of victim services, their treatment by criminal justice professionals, and the processes they experience in the system are more important for their satisfaction than the outcome of their case [85]. Existing good practice should be followed for victims, regardless of the age of the alleged offender.

There is little evidence that victims consistently demand stricter punishments. While some literature suggests that the general public believe that punishments for child offenders should be tougher [86], this opinion is seldom informed by knowledge about the actual likely tariffs for different crimes [87]. Where offenders are unknown to them, stereotypes influence victims’ attitudes towards justice goals and processes [88]. When asked to recommend improvements to the YJS, adults in New Zealand wanted most often wanted change to systemic issues, such as government policy, education, parenting and poverty [87]. Those who had been victims were more likely to express support for preventative measures within schools and families, rather than for increased punishment or a focus on rehabilitation [87].

Restorative justice is valued by victims, particularly when they perceive offences are related to immaturity. Restorative justice is becoming a mainstream response to young people who offend in many countries [89] and there is a wealth of literature on restorative justice theories, principles, processes, practice issues and the experiences of both offenders and victims [90], but very little that considers differences by age of the alleged offender. Evaluations of restorative justice interventions are generally positive, with high levels of victim satisfaction. Securing an apology is a common aim within restorative approaches, although there are concerns that apologies which are required by authority lack sincerity [91] and that victims may feel under pressure to accept the apology [92]. Nonetheless, apologies are valued by many victims, with apologies made by younger children being deemed more sincere than those of older children [93]. Similarly, children and young people are often believed to be more redeemable than adults [88]. Offences are more likely to be seen as related to immaturity, negative peer influence and carelessness rather than deliberate intent, and thus requiring responses based on compassion and understanding–- albeit with longer-term interventions for contributory factors, such as substance misuse [94] [88]. Taken together, this research suggests that restorative approaches with younger children, including those below the current age of criminal responsibility, may be supported by victims and the wider public. That said, for the youngest children there are concerns that younger children may not have the emotional maturity or communication skills needed to fully engage in restorative conversations, and questions about additional skills or training RJ facilitators need to work with pre-adolescent children.

5. Case studies

6. Conclusions and recommendations

Having a minimum age of criminal responsibility of 10 years in England, Wales and Ireland results in contradictions and inconsistencies within the laws of the UK, is out of step with international norms, and contravenes the UN Convention on the Rights of the Child.

There is extensive evidence to demonstrate that criminalisation of younger children has negative consequences for children, including increasing the likelihood of future criminal acts. There is a strong argument for increasing the MACR to 14 years.

This review of literature and practice sought out evidence for the success of non-criminalised responses to harms done by children aged under 14 years. The impact on reoffending, on child outcomes and on victims was considered.

There are few interventions in either the international research literature or in UK practice that are specific to the 8-13 year old age group. In fact, most services and interventions reflect the consequences of the current MACR; since all children between the aged of 10 and 17 years could be held criminally responsible it is rare for age specific distinctions to be made within this. That said, in practice, Youth Offending Services (YOS) individualise their responses so it is unlikely that younger children would be treated in the same way as older children by staff.

Victims of crimes do not report greater satisfaction when children receive criminal sentences. Victims feel most supported when their voices are heard and the harm to them is taken seriously, regardless of the outcome for the offender. One research study in New Zealand suggests that adult victims of child offenders recognise their vulnerability and are concerned for their welfare.

No evidence was found within this review to suggest that raising the MACR would do harm but plenty to suggest that better outcomes would be achieved for children and wider society.

The process of conducting this review has highlighted the need for more data about the involvement of younger children in offending or harmful behaviour and appropriate interventions for them. Currently within the UK, data on these children is subsumed within the reporting of all work by Youth Offending Teams. There is also very limited recording of their involvement in non-criminal justice or diversionary interventions, which limited our ability to know how often these are used and to monitor outcomes.

In countries with higher MACR, data and research about these younger children falls out with the criminal justice system. and is instead included in wider discussions of responding to broader categories of behaviours and problems.

More rigorous data collection on interventions and outcomes for younger children who commit harms and/or more serious offences is needed to inform future practice but this does not preclude raising the minimum age of criminal responsibility to 14.

The authors of this report therefore recommend:

  • Raising the age of criminal responsibility to 14 and transferring responsibility for children aged 10-13 years who cause harm from youth justice to children’s social care.
  • Expand access to systemic therapies, cognitive or behavioural therapies for younger children whose behaviour causes harm.
  • Build a system that recognises and responds to the impact of Adverse Childhood Experiences and Trauma on child behaviour. This is likely to mean providing training for professionals across sectors (including the police). It may also mean introducing earlier assessments to identify children likely to need more intensive support.
  • Avoid using cautions, court orders, or convictions as necessary qualifiers for intensive intervention (for example from mental health teams). This is not appropriate for a criminal justice response to be required before young children receive the support they need. Needs should be assessed, and high-quality intervention considered at the earliest signs of aggressive, antisocial or disruptive behaviour.
  • Where used, police-led diversion should occur pre-arrest, to avoid unnecessary criminalisation. Diversion should be embedded within a strategy which seeks to reduce punitive contacts with the police, and supports the police to link children to other supporting services (e.g. referrals to CAMHS, domestic violence services, referrals to local youth activities).
  • Take a youth-led approach to design and/or selection of activities and interventions. Evidence from practice suggests this will increase uptake and avoid wasting resources on unpopular programmes.
  • Neither area-based curfews, nor schemes which bring younger children into contact with older children with more entrenched offending should be used. Neither are supported by the evidence and are unlikely to be a good use of resources.
  • Develop a culture of cooperation and support within linked Youth Offending services. In case studies, participants reported this facilitated programme success regardless of the approach taken. Good relationships between services improved the experiences of professionals, families and children.

There are also gaps in our knowledge that need filling. We recommend:

  • Recording and monitoring non-criminal justice or diversionary actions; these should be recorded anonymously and reported at aggregate level to avoid appearing on individual children’s records. This would enable monitoring and evaluation of the use of tools such as youth justice orders, injunctions to prevent nuisance and annoyance and out of court disposals.
  • Consider the differential impact of crime reduction interventions according to child age. This is needed both when reporting on the impact of commissioned activities and when commissioning research.
  • Develop and test alternative interventions targeting children aged 8-13 years engaged in offending behaviour.
  • Commission research into models of restorative justice or mediation appropriate for younger children, including when victims and offenders are in close relationships
  • More rigorous data collection on interventions and outcomes for younger children who commit more serious offences

7. Glossary (Incorporating definitions provided by [69,73])

Glossary
Acronym In full Definition
ACE Adverse Childhood Experiences Potentially traumatic negative events that occur in childhood, eg death of a parent, exposure to violence.
AssetPlus AssetPlus A detailed assessment tool developed by the Youth Justice Board (YJB), and used with Level 2 Diversion Programme cases and statutory disposals.
CAMHS Child and Adolescent Mental Health Services  
CBT Cognitive Behavioural Therapy A talking therapy aimed at changing the way you think and behave.
CD Conduct Disorder Repeated and persistent pattern of behaviour showing aggression, destruction of proper, theft or violation of rules.
CJS Criminal Justice System The wide definition of the system of justice in England and Wales, with responsibility for the administration of justice for adult and youth offenders.
FTEs First Time Entrants to the Youth Justice System National Performance Indicator measured against an annually agreed local target. A young person receiving a formal statutory disposal or outcome for the first time counts as a ‘First Time Entrant to the Youth Justice System’.
NFA No Further Action When the police decide not to charge someone with an offence, usually on the grounds of limited evidence or not being within the public interest.
OoCD Out of Court Disposal Comprised of statutory and non-statutory disposals. The non-statutory disposal is a community resolution. The statutory disposals are a Youth Caution and a Youth Conditional Caution.
TRM Trauma Recovery Model  
YOT/YOS Youth Offending Team/ Service Multi-disciplinary team within the local authority, whose remit is to work with children aged 10-17 who are involved in offending behaviour; some YOTs also work with children ‘at risk’ of involvement in offending.
YJS Youth Justice System The system of justice in England and Wales responsible for the administration of youth justice. It includes the police, youth offending teams and the courts.
YJB Youth Justice Board  

 

References

1. United Nations Committee on the Rights of the Child. Concluding observations on the fourth periodic report of the United Kingdom of Great Britain and Northern Ireland. Geneva: United Nations, 2008.

2.  Age of Criminal Responsibility (Scotland) Act 2019, (2019).

3. Brown A, Charles A. The Minimum Age of Criminal Responsibility: The need for a holistic approach. Youth Justice. 2021;12:153-71.

4. Graham J, Perrott S, Marshal K. A Review of the Youth Justice System in Northern Ireland. Belfast: Department of Justice Northern Ireland 2011.

5. Delmage E. The Minimum Age of Criminal Responsibility: A medico-legal perspective. Youth Justice. 2012;13:102-10.

6. Lamb ME, Sim MP. Developmental Factors Affecting Children in Legal Contexts. Youth Justice. 2013;13(2):131-44.

7. Royal College of Psychiatrists. Child Defendants London: Royal College of Psychiatrists, 2006  Contract No.: Occasional Paper 56.

8. Ryan EP, Murrie DC. Competence to Stand Trial and Young Children: Is the presumption of competence valid? Journal of Forensic Psychology Practice. Journal of Forensic Psychology Practice. 2008;5:89-102.

9. UNICEF UK. A RIGHTS-BASED ANALYSIS OF YOUTH JUSTICE IN THE UNITED KINGDOM. 2020.

10. Gray E. What Happens to Persistent and Serious Young Offenders When They Grow Up. London: Youth Justice Board, 2013.

11. Farrington DP, Welsh BC. Saving children from a life of crime: early risk factors and effective interventions. Oxford: Oxford University Press; 2007.

12. McAra L, McVie S. Youth justice?: The impact of system contact on patterns of desistance from offending. European Journal of Criminology. 2007;43:315-45.

13. Goldson B. Unsafe, Unjust and Harmful to Wider Society': Grounds for Raising the Minimum Age of Criminal Responsibility in England and Wales. Youth Justice. 2013;13(2):111-30.

14. Youth Justice Board, Ministry of Justice. Youth Justice Statistics 2019/20. England and Wales. 2021 28 January 2021. Report No.

15. Ministry of Youth Justice Board, Ministry of Justice. Assessing the needs of sentenced children in the youth justice system 2019/20, England and Wales. London:Justice, 2021.

16. Goldson B. Excavating Youth Justice Reform: Historical mapping and speculative prospects. The Howard Journal of Crime and Justice. 2020;59:317-34.

17. Solanki  AR, Bateman T, Boswell G, Hill E. Antisocial Behaviour Orders for Young People. London: Youth Justice Board for England and Wales, 2006.

18. Bateman T. Criminalising children for no good purpose: The age of criminal responsibility in England and Wales 2012.

19. Sutherland A, Disley E, Cattell J, Bauchowitz S. An Analysis of Trends in First Time Entrants to the Youth Justice System. Cambridge: RAND, 2017.

20. Her Majesty’s Inspectorate of Probation. The experiences of black and mixed heritage boys in the youth justice system, . London: HMIP, 2021.

21. Haines K, Case S. Positive Youth Justice, . Bristol: Policy Press; 2015.

22.ONS. Analysis of population estimates tool for UK. Office for National Statistics; 2021.

23. Staines J. Risk, Adverse Influence and Criminalisation Understanding the over-representation of looked after children in the youth justice system. London: Prison Reform Trust, 2016.

24. Caluori J, Corlett M, Stott J. County Lines and Looked After Children. London: Crest Advisory, 2020.

25. Lammy D. The Lammy Review: An Independent Review into the Treatment of, and Outcomes for, Black, Asian and Minority Ethnic Individuals in the Criminal Justice System. London: Lammy Review, 2017.

26. Department for Education. Children accommodated in secure children's homes. London: DfE, available at  https://explore-education-statistics.service.gov.uk/find-statistics/children-accommodated-in-secure-childrens-homes, 2021.

27. Lambie I, Randell I. The impact of incarceration on juvenile offenders. Clin Psychol Rev. 2013;33(3):448-59. Epub 2013/03/05. doi: 10.1016/j.cpr.2013.01.007. PubMed PMID: 23454219.

28. Mulder E, Vermunt J, Brand E, Bullens R, van Marle H. Recidivism in subgroups of serious juvenile offenders: different profiles, different risks? Crim Behav Ment Health. 2012;22(2):122-35. Epub 2012/01/04. doi: 10.1002/cbm.1819. PubMed PMID: 22213477.

29. de Ruigh EL, Popma A, Twisk JWR, Wiers RW, van der Baan HS, Vermeiren R, et al. Predicting quality of life during and post detention in incarcerated juveniles. Qual Life Res. 2019;28(7):1813-23. Epub 2019/03/16. doi: 10.1007/s11136-019-02160-6. PubMed PMID: 30875009; PubMed Central PMCID: PMCPMC6571096.

30. Vinson SY, Waldman RJ. The pandemic paused the US school-to-prison pipeline: potential lessons learned. Lancet Child Adolesc Health. 2020;4(11):799-800. Epub 2020/09/21. doi: 10.1016/s2352-4642(20)30306-0. PubMed PMID: 32950069; PubMed Central PMCID: PMCPMC7498415.

31. Padgaonkar NT, Baker AE, Dapretto M, Galván A, Frick PJ, Steinberg L, et al. Exploring Disproportionate Minority Contact in the Juvenile Justice System Over the Year Following First Arrest. J Res Adolesc. 2021;31(2):317-34. Epub 2020/12/07. doi: 10.1111/jora.12599. PubMed PMID: 33280192; PubMed Central PMCID: PMCPMC8127356.

32. Bowen P. Oral evidence: Children and young people in custody, HC 306. Q188. House of Commons Justice Committee, 2020 30/6/2020. Report No.

33. Ipsos MORI. Supporting Families against Youth Crime Fund Evaluation: Final Report. London: Ipsos MORI, 2021.

34. Scottish Government. Age of Criminal Responsibility (Scotland) Act 2019 Child Interview Rights Practitioners – Code Of Practice. https://www.gov.scot/publications/age-criminal-responsibility-scotland-act-2019-child-interview-rights-practitioners-code-practice/: 2021.

35. Petrosino A, Turpin-Petrosino C, Hollis-Peel ME, Lavenberg JG. Scared Straight and Other Juvenile Awareness Programs for Preventing Juvenile Delinquency: A Systematic Review. Campbell Systematic Reviews. 2013;9(1):1-55. doi: https://doi.org/10.4073/csr.2013.5.

36. Villettaz P, Gillieron G, Killias M. The Effects on Re-offending of Custodial vs. Non-custodial Sanctions: An Updated Systematic Review of the State of Knowledge. Campbell Systematic Reviews. 2015;11(1):1-92. doi: https://doi.org/10.4073/csr.2015.1.

37. Cassidy T, Bowman B, McGrath C, Matzopoulos R. Brief report on a systematic review of youth violence prevention through media campaigns: Does the limited yield of strong evidence imply methodological challenges or absence of effect? J Adolesc. 2016;52:22-6. doi: https://dx.doi.org/10.1016/j.adolescence.2016.07.002. PubMed PMID: 27485240.

38. Cassidy T, Inglis G, Wiysonge C, Matzopoulos R. A systematic review of the effects of poverty deconcentration and urban upgrading on youth violence. Health Place. 2014;26:78-87. doi: https://dx.doi.org/10.1016/j.healthplace.2013.12.009. PubMed PMID: 24412655.

39. Pardini D. Empirically Based Strategies for Preventing Juvenile Delinquency. Child Adolesc Psychiatr Clin N Am. 2016;25(2):257-68. doi: https://dx.doi.org/10.1016/j.chc.2015.11.009. PubMed PMID: 26980128.

40. McDaniel DD, Logan JE, Schneiderman JU. Supporting Gang Violence Prevention Efforts: A Public Health Approach for Nurses. Online J Issues Nurs. 2014;19(1):3. PubMed PMID: 26812200.

41. Daykin N, de Viggiani N, Pilkington P, Moriarty Y. Music making for health, well-being and behaviour change in youth justice settings: a systematic review. Health Promot Internation. 2013;28(2):197-210. doi: https://dx.doi.org/10.1093/heapro/das005. PubMed PMID: 22415559.

42. Rice TR. Violence among young men: the importance of a gender-specific developmental approach to adolescent male suicide and homicide. Int J Adolesc Med Health. 2015;27(2):177-81. doi: https://dx.doi.org/10.1515/ijamh-2015-5008. PubMed PMID: 25411987.

43. Parrish DE. Achieving Justice for Girls in the Juvenile Justice System. Soc Work. 2020;65(2):149-58. doi: https://dx.doi.org/10.1093/sw/swaa005. PubMed PMID: 32280972.

44. Thomann A, Keyes L, Ryan A, Graaf G. Intervention Response to the Trauma-Exposed, Justice-Involved Female Youth: A Narrative Review of Effectiveness in Reducing Recidivism. Int J Environ Res Public Health. 2020;17(20):12. doi: https://dx.doi.org/10.3390/ijerph17207402. PubMed PMID: 33053618.

45. Sugimoto-Matsuda JJ, Braun KL. The role of collaboration in facilitating policy change in youth violence prevention: a review of the literature. Prev Sci. 2014;15(2):194-204. doi: https://dx.doi.org/10.1007/s11121-013-0369-7. PubMed PMID: 23430580.

46. Granski M, Javdani S, Anderson VR, Caires R. A Meta-Analysis of Program Characteristics for Youth with Disruptive Behavior Problems: The Moderating Role of Program Format and Youth Gender. Am J Community Psychol. 2020;65(1-2):201-22. doi: https://dx.doi.org/10.1002/ajcp.12377. PubMed PMID: 31449683.

47. Sawyer AM, Borduin CM, Dopp AR. Long-term effects of prevention and treatment on youth antisocial behavior: A meta-analysis. Clinical Psychology Review. 2015;42:130-44. doi: https://dx.doi.org/10.1016/j.cpr.2015.06.009. PubMed PMID: 26197725.

48. von Sydow K, Retzlaff R, Beher S, Haun MW, Schweitzer J. The efficacy of systemic therapy for childhood and adolescent externalizing disorders: a systematic review of 47 RCT. Fam Process. 2013;52(4):576-618. doi: https://dx.doi.org/10.1111/famp.12047. PubMed PMID: 24102196.

49. Waddell C, Schwartz C, Andres C, Barican JL, Yung D. Fifty years of preventing and treating childhood behaviour disorders: a systematic review to inform policy and practice. Evid Based Ment Health. 2018;21(2):45-52. doi: https://dx.doi.org/10.1136/eb-2017-102862. PubMed PMID: 29703717.

50. Bakker MJ, Greven CU, Buitelaar JK, Glennon JC. Practitioner Review: Psychological treatments for children and adolescents with conduct disorder problems - a systematic review and meta-analysis. J Child Psychol Psychiatry. 2017;58(1):4-18. doi: https://dx.doi.org/10.1111/jcpp.12590. PubMed PMID: 27501434.

51. de Vries SL, Hoeve M, Assink M, Stams GJ, Asscher JJ. Practitioner review: Effective ingredients of prevention programs for youth at risk of persistent juvenile delinquency--recommendations for clinical practice. J Child Psychol Psychiatry. 2015;56(2):108-21. doi: https://dx.doi.org/10.1111/jcpp.12320. PubMed PMID: 25143121.

52. Florean IS, Dobrean A, Pasarelu CR, Georgescu RD, Milea I. The Efficacy of Internet-Based Parenting Programs for Children and Adolescents with Behavior Problems: A Meta-Analysis of Randomized Clinical Trials. Clinical Child & Family Psychology Review. 2020;23(4):510-28. doi: https://dx.doi.org/10.1007/s10567-020-00326-0. PubMed PMID: 32897527.

53. NICE. Antisocial behaviour and conduct disorders in children and young people: recognition and management. Clinical guideline [CG158]. London: National Institute for Health and Clinical Excellence, 2013.

54. Riise EN, Wergeland GJH, Njardvik U, Ost LG. Cognitive behavior therapy for externalizing disorders in children and adolescents in routine clinical care: A systematic review and meta-analysis. Clinical Psychology Review. 2021;83:101954. doi: https://dx.doi.org/10.1016/j.cpr.2020.101954. PubMed PMID: 33418192.

55. Tolan P, Henry D, Schoeny M, Bass A, Lovegrove P, Nichols E. Mentoring Interventions to Affect Juvenile Delinquency and Associated Problems: A Systematic Review. Campbell Systematic Reviews. 2013;9(1):1-158. doi: https://doi.org/10.4073/csr.2013.10.

56. Valdebenito S, Eisner M, Farrington DP, Ttofi MM, Sutherland A. School-based interventions for reducing disciplinary school exclusion: a systematic review. Campbell Systematic Reviews. 2018;14(1):i-216. doi: https://doi.org/10.4073/csr.2018.1.

57. Wilson DB, Brennan I, Olaghere A. Police-initiated diversion for youth to prevent future delinquent behavior: a systematic review. Campbell Systematic Reviews. 2018;14(1):1-88. doi: https://doi.org/10.4073/csr.2018.5.

58. Wilson DB, Gill C, Olaghere A, McClure D. Juvenile Curfew Effects on Criminal Behavior and Victimization: A Systematic Review. Campbell Systematic Reviews. 2016;12(1):1-97. doi: https://doi.org/10.4073/csr.2016.3.

59. Stein DM, Deberard S, Homan K. Predicting success and failure in juvenile drug treatment court: a meta-analytic review. J Subst Abuse Treat. 2013;44(2):159-68. doi: https://dx.doi.org/10.1016/j.jsat.2012.07.002. PubMed PMID: 22980448.

60. Strang H, Sherman LW, Mayo-Wilson E, Woods D, Ariel B. Restorative Justice Conferencing (RJC) Using Face-to-Face Meetings of Offenders and Victims: Effects on Offender Recidivism and Victim Satisfaction. A Systematic Review. Campbell Systematic Reviews. 2013;9(1):1-59. doi: https://doi.org/10.4073/csr.2013.12.

61. Loy JH, Merry SN, Hetrick SE, Stasiak K. Atypical antipsychotics for disruptive behaviour disorders in children and youths. Cochrane Database Syst Rev. 2017;8:CD008559. doi: https://dx.doi.org/10.1002/14651858.CD008559.pub3. PubMed PMID: 28791693.

62. Pringsheim T, Hirsch L, Gardner D, Gorman DA. The pharmacological management of oppositional behaviour, conduct problems, and aggression in children and adolescents with attention-deficit hyperactivity disorder, oppositional defiant disorder, and conduct disorder: a systematic review and meta-analysis. Part 2: antipsychotics and traditional mood stabilizers. Can J Psychiatry. 2015;60(2):52-61. PubMed PMID: 25886656.

63. Pringsheim T, Hirsch L, Gardner D, Gorman DA. The pharmacological management of oppositional behaviour, conduct problems, and aggression in children and adolescents with attention-deficit hyperactivity disorder, oppositional defiant disorder, and conduct disorder: a systematic review and meta-analysis. Part 1: psychostimulants, alpha-2 agonists, and atomoxetine. Can J Psychiatry. 2015;60(2):42-51. PubMed PMID: 25886655.

64. Early Intervention Foundation. EIF evidence standards London: Early Intervention Foundation,; 2017 [cited 2022]. Available from: https://guidebook.eif.org.uk/eif-evidence-standards.

65. McEwan D. Diverting children and young people from Serious and Organised Crime (SOC): Guidance for Practice on Recognising and Preventing the Exploitation of Children by SOC. Glasgow: Centre for Youth & Criminal Justice, 2018.

66. Dawson H, Dunn S, Morgan V. AN EVALUATION OF THE EARLY INTERVENTION PROGRAMME FOR THE PREVENTION OF OFFENDING. Belfast: INDEPENDENT RESEARCH SOLUTIONS, 2010.

67. Nee C, Ellis T, Morris P, Wilson A. Addressing criminality in childhood: is responsivity the central issue? Int J Offender Ther Comp Criminol. 2013;57(11):1347-73. Epub 2012/10/17. doi: 10.1177/0306624x12460377. PubMed PMID: 23070956.

68. YJB. Evaluation of Preventing Youth Offending Project (PYOP): Youth Justice Resource Hub. Available from: https://yjresourcehub.uk/evaluation-library/item/339-evaluation-of-preventing-youth-offending-project-pyop.html.

69. Cordis Bright. Evaluation of the Enhanced Case Management approach. Cardiff: Government Social Research, 2017.

70. Shaw B, Bernardes E. Forging futures through mentoring. A risk worth pursuing? London: Children's Commissioner for England, 20018.

71. Walker J, Thompson C, Laing K, Raybould S, Coombes M, Procter S, et al. Youth Inclusion and Support Panels: Preventing Crime and Antisocial Behaviour? Newcastle: Newcastle Centre for Family Studies, 2007  Contract No.: DCSF-RW018.

72. YJB. Youth Justice Liaison and Diversion Scheme: YJB Youth Justice Resource Hub;  [cited 2022 31/3/2022]. Available from: https://yjresourcehub.uk/evaluation-library/item/346-youth-justice-liaison-and-diversion-scheme.html.

73. Tyrrell K, Bond E, Manning M, Dogaru C, D'Cruz C. Diversion, Prevention and Youth Justice: a model of integrated decision making An evaluation of the Suffolk Youth Offending Service Diversion Programme. Ipswich: University of Suffolk, 2017.

74. Haines A, Goldson B, Haycox A, Houten R, Lane S, McGuire J, et al. Evaluationof the Youth Justice Liaison and Diversion (YJLD) Pilot Scheme. Final Report. University of Liverpool. 2012.

75. Robin-D’Cruz C, Tibbs E. Mapping youth diversion in England and Wales. London: Centre for Justice Innovation.

76. Mason C, Downward P, Walpole C, Cleland J, Aldridge J. Youth Crime Reduction and Sport Pilot Project. Evaluation Report. Loughborough Loughborough University, 2017.

77. Barrett KL, Jennings WG, Lynch MJ. The relation between youth fear and avoidance of crime in school and academic experiences. Journal of School Violence. 2012;11(1):1-20. PubMed PMID: 2012-00092-001.

78. Hayden C, Dlugosz G. Secondary school children and the experience of robbery: A survey in three south London schools. Crime Prevention and Community Safety. 2012;14(2):122-39. PubMed PMID: 2012-09512-002.

79. Howard League. Children as Victims. Child-sized crimes in a child-sized world. London: Howard League for Penal Reform, 2007.

80. Gee KA, Cooc N. Will I be victimized at school today? How schools influence the victimization experiences of Asian American teenagers. Asian American Journal of Psychology. 2019;10(4):316-25. PubMed PMID: 2019-17532-001.

81. Hughes MR, Gaines JS, Pryor DW. Staying away from school: Adolescents who miss school due to feeling unsafe. Youth Violence and Juvenile Justice. 2015;13(3):270-90. PubMed PMID: 2015-24824-005.

82. Stripe N. The nature of violent crime in England and Wales: Year ending March 2020. . London: Office for National Statistics, 2021.

83. Hunter C, Piper C. Parent Abuse: Can Law Be the Answer? Social Policy and Society. 2012;11(2):217-27. doi: http://dx.doi.org/10.1017/S1474746411000637. PubMed PMID: 952708720.

84. Nixon J. Practitioners’ Constructions of Parent Abuse. Social Policy and Society. 2012;11:229-39.

85. Wedlock E, Tapley J. What works in supporting victims of crime: A rapid evidence assessment. London: Victims' Commissioner's Office, 2016.

86. Russell N, Morgan R. Sentencing of domestic burglary (Sentencing Advisory Panel Research Report 1). London: Home Office Cited by Barretto C, Miers S, Lambie I. The Views of the Public on Youth Offenders and the New Zealand Criminal Justice System. Int J Offender Ther Comp Criminol. 2018;62(1):129-49. 2001.

87. Barretto C, Miers S, Lambie I. The Views of the Public on Youth Offenders and the New Zealand Criminal Justice System. Int J Offender Ther Comp Criminol. 2018;62(1):129-49. doi: 10.1177/0306624x16644500. PubMed PMID: WOS:000418185700007.

88. Paul GD. The Influence of Belief in Offender Redeemability and Decision-Making Competence on Receptivity to Restorative Justice. Negot Confl Manag Res. 2021;14(1):1-20. doi: 10.1111/ncmr.12176. PubMed PMID: WOS:000685167600001.

89. Reyes-Quilodran CA, LaBrenz CA, Donoso G. A path to peace: A comparative study of practitioners' views of juvenile restorative justice in Chile, England, Italy, and Sweden. International Social Work. 2019;62(1):255-70. PubMed PMID: 2018-66484-020.

90. Bolitho JJ. Restorative Justice: The Ideals and Realities of Conferencing for Young People. Crit Criminol. 2012;20(1):61-78. doi: 10.1007/s10612-011-9150-z. PubMed PMID: WOS:000306244100005.

91. Choi JJ, Severson M. "What! What kind of apology is this?": The nature of apology in victim offender mediation. Children and Youth Services Review. 2009;31(7):813-20. doi: 10.1016/j.childyouth.2009.03.003. PubMed PMID: WOS:000267099300013.

92. Choi JJ, Green DL, Kapp SA. A qualitative study of victim offender mediation: Implications for social work. Journal of Human Behavior in the Social Environment. 2010;20(7):857-74. PubMed PMID: 2010-23636-002.

93. Allan A, de Mott J, Larkins IM, Turnbull L, Warwick T, Willet L, et al. The impact of voluntariness of apologies on victims' responses in restorative justice: findings of a quantitative study. Psychiatr Psychol Law.17. doi: 10.1080/13218719.2021.1956383. PubMed PMID: WOS:000711811000001.

94. Van Camp T. Understanding victim participation in restorative practices: Looking for justice for oneself as well as for others. European Journal of Criminology. 2017;14(6):679-96. doi: 10.1177/1477370816682981. PubMed PMID: WOS:000415832800003.

95. Office for National Statistics. The nature of violent crime in England and Wales: Year ending March 2020. A summary of violent crime from the Crime Survey for England and Wales and police recorded crime. London: ONS 2021.