Psychological Resilience Interventions to Reduce Recidivism in Young

ANGOR UNIVERSITY People: A Systematic Review Hodgkinson, Rowan; Beattie, Stuart; Roberts, Ross; Hardy, Lewis

Adolescent Research Review

DOI: https://doi.org/10.1007/s40894-020-00138-x

PRIFYSGOL BANGOR / B E-pub ahead of print: 24/07/2020

Publisher's PDF, also known as Version of record

Cyswllt i'r cyhoeddiad / Link to publication

Dyfyniad o'r fersiwn a gyhoeddwyd / Citation for published version (APA): Hodgkinson, R., Beattie, S., Roberts, R., & Hardy, L. (2020). Psychological Resilience Interventions to Reduce Recidivism in Young People: A Systematic Review. Adolescent Research Review. https://doi.org/10.1007/s40894-020-00138-x

Hawliau Cyffredinol / General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal ?

Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

30. Sep. 2021 Adolescent Research Review https://doi.org/10.1007/s40894-020-00138-x

SYSTEMATIC REVIEW

Psychological Resilience Interventions to Reduce Recidivism in Young People: A Systematic Review

Rowan Hodgkinson1 · Stuart Beattie1 · Ross Roberts1 · Lew Hardy1

Received: 30 April 2020 / Accepted: 14 July 2020 © The Author(s) 2020

Abstract “Diversion” schemes encouraging children and young people away from ofending have successfully reduced the numbers of young people within the youth justice system. However, for those not successfully diverted, recidivism remains obstinately high. Many of those remaining in the youth justice system appear to have complex psychological needs. Research has also shown that many of this group have experienced a high number of adverse childhood experiences. Investigation into the potential consequences of these experiences suggests the potential disruption of normative adolescent psychological growth. Domains may include emotional, cognitive, behavioral, and interpersonal development. This review assesses the efective- ness of individual interventions that had a psychological focus and succeeded in reducing recidivism. A systematic research review from 2000 to 2019 yielded 206 studies for youth ofenders, and of these, 14 met the criteria for inclusion. Sample size varied greatly, from 30 to 3038. Research design, follow-up period and intervention content also varied greatly. Further, intervention success for recidivism ranged from almost total desistance to changes (increased time to re-ofend) afecting only 50% of the intervention group. Psychological changes as a result of intervention included an increased sense of coher- ence, improved emotion recognition, more positive decision-making and reduced defance. However, none of the studies conducted follow-up psychological assessments post-intervention. Although youth crime is a priority for policy makers, so far research has fallen short of fully examining how the development of psychological resilience via interventions may help reduce persistent ofending.

Keywords Resilience · Psychological interventions · Recidivism · Re-ofending · Juvenile · Youth · Delinquency · Systematic review

Introduction down” by less serious or “adolescent-limited” youth ofend- ers (Moftt 2003). This phenomenon has highlighted the Due to the overall success of youth justice diversion schemes complex needs of those individuals who remain in the youth such as counselling and victim awareness, the pool of young justice system, many of whom become repeat ofenders. people within the youth justice system is no longer “watered Research shows that this population experience high levels of trauma, difculties understanding emotional states, sub- stance misuse, and mental health issues (Mallet and Tedor This research is part of a Knowledge Economy Skills Scholarships 2018). Re-ofending rates are high, suggesting a distinct (KESS 2) funded research project, project code: C80815. KESS 2 is a major pan-Wales operation supported by European Social need for focused research attention on this worldwide con- Funds (ESF) through the Welsh Government. KESS 2 links cern. However, despite the notable psychological difculties companies and organisations with academic expertise in the experienced by those who re-ofend, there has not been a Higher Education sector in Wales to undertake collaborative systematic review of psychological interventions that have research projects, working towards a PhD or Research Masters qualifcation. The collaborating company on this project is the succeeded in reducing re-ofending. This study, therefore, Youth Justice Service for Gwynedd and Ynys Môn to which none aims to address that research gap by identifying successful of the authors have any afliation or fnancial connections. interventions world-wide and reporting on the psychological changes occurring along with reduced re-ofending. * Rowan Hodgkinson [email protected] Extended author information available on the last page of the article

Vol.:(0123456789)1 3 Adolescent Research Review

Defnition of “Young Ofender” Psychological Development

A “young ofender” is a child or a youth convicted or cau- Adolescence is a crucial period when cognitive and emo- tioned by the police. The age of criminal responsibility var- tional skills develop for successful transition into adult- ies between countries. In Europe and Asia, it is 12–14. In hood (Wood et al. 2018). However, research has shown the UK, Ireland and Australia it is 10. However, in the USA, that children and youth in the criminal justice system are 33 out of the 50 states set no minimum age but instead apply more likely to experience delayed cognitive development, a capacity test. Of the 17 that do, North Carolina has the evidenced by factors such as poor emotional regulation lowest at 7, and Wisconsin has the highest at 10. In Scandi- and low academic attainment (Wolf and Baglivio 2017). navian countries which include Denmark, Finland, Norway, Consequently, this may lead to low levels of psychologi- and Sweden, the legal age of criminal responsibility is 15. In cal resilience, e.g. inability to successfully overcome dif- most countries, youth justice services can work with individ- fculties. Further, measures of re-ofending and resilience uals until they turn 18 (although the United Nations class a do not always go hand in hand, meaning the relation- “youth” as between the ages of 15 and 24, Factsheet on Juve- ship between psychological development/resilience and nile Justice 2008). In the UK, a “prolifc young ofender” is reduced ofending is unclear (cf. Daykin et al. 2017). an individual aged between 10 and 17 who has more than 25 One cause of delayed cognitive development in young separate ofences (Johns et al. 2018). people is the experience of trauma. For example, studies Global Youth Justice report that that over the last 5 years, report that these individuals often have difculties in rec- the most common reasons for convictions/cautions of chil- ognizing emotions in others, or in identifying and describ- dren and youth worldwide consisted of theft, vandalism, ing their feelings (alexithymia) in themselves (Möller et al. underage drinking, disorderly conduct (e.g., fghting or 2014). An inability to recognize emotions may be one of assault), marijuana possession, underage smoking, curfew the ways in which traumatic experiences impede norma- violation, school disciplinaries and trafc violations (e.g., tive psychological development (Eichhorn et al. 2014). underage driving) (Top 25 crimes, ofences and violation Delayed cognitive development such as language impair- 2018). While youth diversion schemes are efective at ensur- ment has also been specifcally related to children and ing a high number of children and youth do not go on to youth who ofend (Snow et al. 2015). Overall, research commit further or more serious crime, they are not always fnds that experiencing trauma at a young age often leads successful (Wilson and Hoge 2013). For example, a 2015 to higher levels of negative emotionality such as anger, report from the U.S.A. compiled data from the 39 states greater levels of anxiety and depression, and low levels of that track recidivism found that 76% of frst-time ofenders relatedness and self-concept (Gibson and Clarbour 2017). re-ofended within 3 years, and 84% within 5 years (MST Psychological trauma can result from adverse childhood Services 2018). In the U.K., re-ofending rates are similar, experiences such as loss of a parent or experiencing vio- especially for youth leaving secure institutions where over lent events, as well as responses to chronic or repetitive two thirds reofend within 12 months of release (Youth jus- experiences such as child abuse, neglect, urban violence, tice facts and fgures 2020). violent relationships, and chronic deprivation (Committee “Serious” crimes committed by children and youth are on Child Maltreatment Research, Policy, and Practice for often associated with gang involvement, e.g. knife crime or the Next Decade 2014). Sudden changes such as loss or violent crime (Association of Directors of Children’s Ser- bereavement may also include having to leave the family vices 2019). “Less serious” ofenders are those that have home due to confict, abuse, or overcrowding (Diaz 2005). committed ofences considered to be non-violent in nature Incarceration also signifes sudden loss of the familiar, a such as property crimes (Turner 2015). Those who commit potentially traumatic experience related to separation pain serious crimes are less likely to receive diversion options (Armstrong and Weaver 2013). Research has also shown such as restorative justice schemes. There are many crimes that the combination of trauma previously experienced other than violent listed as “serious” in various legislations by many of those incarcerated with further deprivation worldwide but these are more likely to be committed by experienced as part of the prison environment, can lead to adults, e.g. drug trafcking, people and arms trafcking, further traumatization (Armour 2012). Therefore, incar- prostitution and child sex ofences, armed robbery, bribery, ceration for youth who may already lack resilience may computer misuse ofences and environment ofences (Seri- further delay psychological growth, hindering normative ous Crime Act 2007). adolescent development and potentially contributing to difculties transitioning to adulthood. General strain theory may help explain why negative childhood experiences lead to ofending behavior in some

1 3 Adolescent Research Review individuals (Agnew 2001). This predicts that delinquent rather than external protective factors (Rutter 2006). That behavior occurs when there are disconnections between is, the extent to which youth develop psychologically dur- common goals and the availability of legitimate ways of ing the critical period of adolescence, may be paramount to reaching those goals. Goals may include desire for mate- how well they are able to create and take advantage of pro- rial items or need for status. For example, deviant subcul- tective factors in times of adversity (Steinberg et al. 2004). tures may arise from the need for social recognition (Barry Given the complex psychological profles of children and 2006). Gang membership provides an achievable means youth who ofend, understanding the efectiveness of psy- of meeting this need (U. S. Department of Justice 2015). chological interventions aimed at reducing ofending in this One study examined the efect of eight strain factors on population is paramount. While studies increasingly fnd that delinquency including both general and specifc factors. individuals can develop these resources at any stage in life, Results showed negative relationships with adults and research generally fnds antisocial individuals tend to have a parental fghting combined with other negative life events better response to intervention in early developmental stages and life hassles were signifcantly associated with delin- such as adolescence (Salekin 2015). quency (Agnew and White 1992). Other studies suggest youth who have experienced childhood adversity are more likely to experience frustration or difculties dealing with Current Study emotions (i.e., negative emotionality) which often mani- fests into aggressive behavior (Wolf and Baglivio 2017). Several researchers have conducted reviews on the efcacy In other words, without necessary skills to manage emo- of various interventions for children and youth engaging in tions or achieve goals/basic needs via conventional chan- delinquent behavior. For example, studies show that factors nels, youth may utilize unhelpful methods to meet these determining intervention success included intervention type, needs such as through violence and/or gang membership. methodological rigor, intervention design, demographics, extent of supervision and intervention philosophy (Lip- Development of Psychological Resilience sey 2009). Other research fnds psychosocial interventions that reduced aggressive and violent ofending were efec- However, not all children and youth who experience signif- tive providing they contained elements of emotional self- cant childhood trauma will go on to engage in anti-social management and focused on increasing interpersonal and behavior. Research suggests that the concept of psychologi- social problem-solving skills (McGuire 2008). However, cal resilience and its development during adolescence, may McGuire (2008) stopped short of outlining exactly how these serve as a protective factor in those who experience trauma variables related to reduced ofending. Therefore, despite that do not ofend (Agaibi and Wilson 2005). The concept research evidence indicating that efective psychological of psychological resilience followed in this review comes interventions reduce delinquent behavior and that psycho- from a theoretical model incorporating stress, emotions, and logical resilience may ofer protection from the adverse behavior whereby processes of belief, appraisal, and coping efects of trauma, there is a need to systematically review mediate the stress responses arising from the individual’s this evidence to better inform researchers and practitioners environment. This in turn can lead to positive or negative of best practice. responses, feeling states, and outcomes (Fletcher and Scott 2010). A chronically negative response such as anger or aggression may indicate low psychological resilience. Posi- Method tive emotional states can also act as a moderating attrib- ute, infuencing the extent to which trauma afects behavior Search Strategy (Infurna et al. 2015). Within the literature, researchers often defne resilience This review followed guidelines from the Preferred Report- as the interplay between risk and protective factors (e.g., ing Items for Systematic Reviews and Meta-analyses State- Stoddard et al. 2013). Although many defnitions of resil- ment (PRISMA, Moher et al. 2009) which enabled sys- ience exist (e.g., Luthar et al. 2000), perhaps most pertinent tematic selection of studies for this review. In July 2019, to children and youth who ofend is that those deemed resil- databases searched consisted of Psycinfo and ASSIA (via the ient have “good psychological functioning and good behav- Proquest platform), PubMedCentral, Wiley, Taylor and Fran- ioral outcomes despite adverse circumstances expected to cis, JSTOR, Cochrane Central, Sage and PsycNet (APA). jeopardize normative growth and adaptation” (Mukherjee These databases allowed full text searches, enabling iden- and Kumar 2017, p. 3). Findings from resilience research tifcation of articles omitting key words in their titles and acknowledge that resistance to adversity may derive from abstracts. The frst author initially read and checked all titles a range of physiological or psychological coping processes and abstracts against the eligibility and exclusion criteria

1 3 Adolescent Research Review listed below. Focusing on psychological interventions for individual’s life) and did not utilize psychological measures, repeat ofenders that had reduced ofending and included as it would not be possible to surmise what psychological psychological measures, searches included studies published changes occurred along with reduced ofending. Other stud- between 2000 and 2019 and written in the English language. ies excluded were those for sex ofenders due to the specifc Initial specifc search terms identifed empirical research nature of this ofence and the large number of interventions on psychological interventions for children and youth who for this population suggests a separate review. ofend, and further search terms emerged during the iterative Following examination of the effect of attrition rate searching process. The search terms are listed as follows: (i.e., individuals not completing a program) and in line (adolescen* OR youth OR young OR teen* OR juvenile with Evans-Chase and Zhou 2014), the authors decided OR ofend* OR persistent OR conduct OR delinquen* OR to exclude studies with attrition of 40% or higher. This is problem) AND (intervention OR program* OR treat* OR because attrition rates may bias outcome evaluations as measure OR outcome OR evaluation) AND (resilien* OR participants may have lacked motivation to engage with protective OR cognitive OR self-regulation OR self-efcacy interventions, reducing their success (Hatcher et al. 2012). OR strengths) AND (individual OR self OR behav* OR Figure 1 shows the PRISMA fow chart detailing the review psychological) AND (recidiv* OR desist* OR justice OR and selection process. re-ofend*). Full-text database searches were key to identify- ing relevant articles as this enabled the inclusion of articles Assessment of Study Quality where key words did not appear in the title or the abstract. The first author followed up the database searches There were notable variations in group sizes, intensity, with backwards and forwards reference searches to identify follow-up periods, and reported outcomes. Therefore, a further relevant articles. The backwards searches involved meta-analysis was not possible. Instead, the 16-item qual- scanning reference lists for further eligible studies, and the ity assessment tool assessed overall study quality in studies forwards searches used the “cited by” function provided by meeting eligibility and inclusion criteria (QATSDD; Sirri- the database used. Documents searched consisted of reviews, yeh et al. 2012). The QATSDD contains a list of criteria for systematic reviews and meta-analyses identified during quantitative and qualitative studies rated on a 4-point scale, the database searches, as well as the intervention studies ranging from 0 (not at all) to 3 (complete). Relevant crite- selected for the review. In addition, the authors conducted ria’s i.e. those applying to quantitative, qualitative, or mixed searches in several key organizations’ websites (Youth Jus- method designs included “Clear description of research set- tice Board, Home Ofce). ting”; “Detailed recruitment data”; “Strengths and limita- tions critically discussed”. Division of the total score for Inclusion and Exclusion Criteria each study by the maximum possible score resulted in a per- centage for standardization purposes. The frst three authors The current review included evaluations of: (a) psychologi- assessed study quality. Specifcally, each author scored ten cal interventions for children and youth who ofend or sys- research studies (i.e., two authors assessed each study). Sub- tematic interventions that included measures of psychologi- sequent discussions resolved any disagreements. Table 1 cal change or targeted specifc psychological variables; (b) presents all studies reviewed along with quality scores. journals written in the English language; (c) children and youth who have ofended more than once or are experienc- Reliability ing a disproportionately high risk for doing; (d) interven- tions that included a control group; (e) interventions that Following the screening of titles in 12 database searches demonstrated a reduction in the amount of ofending, time conducted by the frst author, the second author re-screened between ofending episodes, or a reduction in the serious- 6 randomly selected database searches (totaling 4279 titles) ness of ofending; (f) papers published from the year 2000. to assess interrater reliability. This process did not identify The reason for this selection was that interventions after the any additional titles. The frst author next read the abstracts year 2000 had to meet the frst set of National Standards of all titles identifed and selected 181 for full-text analysis. specifc to the Youth Justice Board for England and Wales The frst author then selected 99 titles for full-text analysis. (a government body introduced under the Crime and Dis- To ensure agreement on the decision-making process, the order Act (1998) to oversee the youth justice system). Its second author read 40 randomly selected article abstracts overall aim is to prevent children and youth from ofending from the 181 selected and specifed which articles they or re-ofending, to ensure custody is safe and secure, and would select for full-text analysis. Subsequent discussions to address underlying causes of ofending behavior (Youth resolved any disagreements. Finally, the frst and second Justice Board 2018). Studies excluded were those which authors read all 99 articles, to reach full agreement on which were systemic in nature (i.e., targeted many areas of the studies to include in the review based on the inclusion and

1 3 Adolescent Research Review

Fig. 1 PRISMA fow chart PRISMA Flow Chart

Records identified through database Additional records identified through searching (n = 47,537) other sources (n = 25) Identification

Records after duplicates removed (n = 47,554)

Screening Titles screened Records excluded (n = 11,658) (n = 11,452)

Full-text articles excluded (n = 192)

Eligibility Full-text articles assessed for eligibility Re-offending not (n = 206) measured – 51 For adults- 31 No control – 27 Family intervention – 13 1st time / at risk – 14 Non-significant – 13 Whole school – 7 Studies included in Study protocol only – 6 synthesis Dropouts as control – 5 Included (n = 14) Systemic/no psych measures – 5 Insufficient detail - 5 Description only – 3 Not an intervention – 2 Comparison of two interventions - 4 Did not control for group differences – 2 Review - 2 Follow-up study – 1 Iatrogenic - 1

exclusion criteria reported above. This agreement consisted and second authors. Subsequently, fourteen interventions of verifying through discussion whether the article fully met met the inclusion criteria. These fourteen studies report on all criteria. twelve diferent interventions (four studies assessed the same intervention twice with diferent sample; Bahr et al. 2015; Burraston et al. 2014; Caldwell and Rybroek 2001; Caldwell Results et al. 2006). The 14 studies included in the review have an asterisk in the bibliography. The database search in July 2019 identifed 47,537 records. Reference and citation searches identified a further 25 Study Design records. After removing eight duplicates and screening a total of 11,658 titles, the frst author read 206 abstracts. Full A total of thirteen studies used quantitative approaches text was next assessed in 99 research studies by both frst and one used mixed methods (Burraston et al. 2014). With

1 3 Adolescent Research Review Quality score (0– 100%) 78 67 83 - expres facial recognize ability to increased sions while the group the remained control same or worsened (violent) re-ofences lated with less severe tion group improved emotion control control emotion tion group improved and less anti-staf; the group intervention to in relation in self-concept increased group; control of crime, justifcation group in impulsivity, them a living owed and that society = 6.42, squared p = 0.011); least efective 33.3%, violent ofences (predicted for actual 28.9%) - interven the multi-week for post-measure of the other three any for tion group but not groups - interven the multi-week for post-measure of the other three any for tion group but not groups Key fndings Key Intervention group participantsIntervention signifcantly corre emotions - facial recognize to Ability - interven signifcant diferences; Borderline the intervention for signifcant increases ofences ‘other’ for efective Most 60.5%, actual 31.1%; chi- (predicted SOC increased signifcantly from pre- to to pre- from signifcantly SOC increased to pre- from decreased PICT signifcantly - tion Measure (Bowen (Bowen tion Measure 2013a , b ) et al. Questionnaire (ECQ; Questionnaire and Masters Roger and 1997 ; Roger 1989 ) Najarian ment Questionnaire Thornton (CAQ; 1987 ) of Criminal Inventory Thinking Styles Walters (PICTS; 1995a , b 1996 ): ence (SOC-13) 1991 ) (Antonovsky of Criminal Thinking - Wal (PICTS, Styles ters 2002 ) Psychological measures measures Psychological utilized - Recogni Emotion Facial The Emotion Control Control The Emotion - Adjust The Custodial The Psychological Sense of Coher Psychological Inventory Inventory Psychological 0.015) level = reduced for both intervention and both intervention for reduced group control reduced of re-ofence severity showed and this group only intervention for signifcant at the (p = 0.04) level was predicted groups as had re-ofended and control predicted tion group reduced re-ofending, and tion group re-ofending, reduced signifcant at the ( p < 0.0001) this was level re-ofending reduce cantly groups with the from multi-week the reducing convictions, previous 17 number to groups as their control SOC/ 1-week change PICT did not Reductions in ofending Reductions 6-month follow-up; volume of ofending volume 6-month follow-up; baseline diferences for Adjusting 1-year ofending control group; ofending control 1-year 56.1% Predicted: 55.1% Actual: about the same as reconvictions Actual both intervention follow-up 2-year At 1-year ofending intervention group; ofending intervention 1-year 47.2% Predicted: 34.7% Actual: This is signifcant at the (p - interven 6–24 months multi-week later; signif - group did not control Multi-week included participants only The analysis and intervention 1-week data for No intervention which teaches teaches which intervention facial recognize to youth expressions enhanced thinking skills, style outward-bound daily expedition, camping - train physical military-style and a work ing, life-skills, placement life ambitions, challenging ambitions, challenging life criminal ideas, cognitive - increas therapy, behavioral problem ing empathy, and advantages solving, of crime, disadvantages communication and social skills training Description of intervention 2-week computerized 2-week 25 weeks containing 5 parts;25 weeks Over 9–30 weeks; promoting promoting 9–30 weeks; Over pre- and post-test and post-test pre- group matched design line diferences pre- and post-test and post-test pre- group matched design line diferences pre- and post-test and post-test pre- group design Study design Study Quasi-experimental base - for Controlled Quasi-experimental base - for Controlled Quasi-experimental (SD) (mean/SD given) not 12.3 non- white (SD) navian European, Mean age Mean age 16.08 (1.2) 16.35 (1.2) Ethnicity reported Not range Age 18–21 18–21 Ethnicity 87.7 white, Mean age Mean age 16.9 (1.1) 18.2 (2.2) Ethnicity 86.6 Scandi - 6.7 East 6.7 African 250 size: n = 50 size: n = intervention size: n = 31 also a 1-week - interven tion and a control 1-week (n =30), but no changes in noted were this group, so not were they included in by the analysis Lindblom et al. ( 2017 ) Participant information M = Male; F = Female Sample size Sample Intervention M = 24 Control M = 26 sample Total Sample size Sample Intervention M = 125 Control M = 125 sample Total Sample size Sample Multi-week Multi-week M/F = 15/2 Multi-week control M/F 13/1 sample Total NB there was Summary of interventions ( 2015 ) et al. et al. ( 2002 ) et al. et al. ( 2017 ) 1 Table Author, year, year, Author, location Hubble et al. Hubble et al. Farrington Lindblom

1 3 Adolescent Research Review Quality score (0– 100%) 79 to post-test indicating a reduced risk indicating a reduced post-test of to the group, intervention for re-ofending man - coping and stress better suggesting agement Key fndings Key Signifcant between group diference pre-test group pre-test diference Signifcant between of Violence Risk of Violence (SAVRY, in Youth Borum 2002 ) et al. Psychological measures measures Psychological utilized Structured Assessment < 0.05) level = less crimes, and less general signif - level, cant at the ( p = < 0.001) level crimes crimes and general remained signifcant at the (p Reductions in ofending Reductions 2 years later; intervention group intervention later; 2 years with groupcompared committed control violent crimes and less property crimes, signifcant at the ( p = < 0.05) in violent reductions later only 3 years for Re-ART participants was participants was Re-ART for ( SD = 23.46). 46.86 weeks in motivation, training Core aggression, self-control optional and group work, stress, modules (reducing impulses, controlling interpretation of events, confict emotions, regulating and a systemic management module). Re-ART family and also included drama mindfulness classes Description of intervention Average intervention length intervention Average pre- and post-test and post-test pre- group design, for controlling using diferences and ANCOVA regression Cox analysis Study design Study Quasi-experimental (SD) can nam Antillean western” - non-west ern” Mean age Mean age 17 (1.2) 16.63 (1.3) Ethnicity 29.5% Dutch - 23.8% Moroc 15.2% Suri - 6.7% Turkish 6.7% Dutch 2.9% “other: 15.3% “other: M/F = 59/4 size: n = 91 Participant information M = Male; F = Female Sample size Sample Intervention Intervention Control M/F = 20/8 sample Total (continued) et al. et al. ( 2018 ) 1 Table Author, year, year, Author, location Hoogsteder Hoogsteder

1 3 Adolescent Research Review Quality score (0– 100%) 73 - derline signifcant level with re-ofending signifcant level derline dissatisfed, and low ‘being unhappy, was self-esteem’ correlated ofending with subsequent Key fndings Key Previous history of re-ofending signifcantly history signifcantly of re-ofending Previous The only risk correlate to variable The only at a bor Youth Justice Service Justice Youth database measuring - envi psychological, ronmental, historical and social risks Psychological measures measures Psychological utilized Projects own national own Projects - ences between the groups re-ofending ences between the those who received However, rates. to longer signifcantly took diversion of the sites (580 days in two re-ofend 84 days vs and 220 days 334 days vs respectively) Reductions in ofending Reductions 15–30 months later, no statistical difer no statistical 15–30 months later, or developmental problems. problems. or developmental Those included entered January 2009 and between 2010, but length of March specifed not intervention special - access to improved ized services as the such Child and Adolescent Mental Health Services to and/or referral (CAMHS) statutory or other relevant voluntary agencies Description of intervention Youth with mental health and/ Youth provided Intervention matched group matched design mod - multi-level elling approach Study design Study Quasi experimental using TREND (SD) white/mixed - other Euro pean Mean age Mean age 14.41 (1.82) 15.67 (1.52) 14.81 (1.38) 15.95 (1.10) 14.60 (1.59) 15.67 (1.40) 14.84 (1.83) 15.79 (1.25) Ethnicity 69.6% white 20.1% non- 10.3% other/ - - tion vention vention M/F = 18/14 vention vention M/F = 86/14 tion size: n = 408 Participant information M = Male; F = Female Sample size Sample - Site 1 Interven M/F = 31/21 Site 1 Control M/F = 31/21 Site 2 Inter Site 2 Control M/F = 20/12 Site 3 Inter Site 3 control M/F = 80/15 - Site 4 Interven M/F = 12/17 Site 4 Control M/F = 9/7 sample Total (continued) ( 2015 ) 1 Table Author, year, year, Author, location Haines et al. Haines et al.

1 3 Adolescent Research Review Quality score (0– 100%) 71 51 56 cally signifcant in each of the six areas signifcant in each cally ( p < 0.05) measured available for 46 of the group. intervention for available of 17.3% in risks a reduction But did show the YLS/CMI by and needs as measured that revealed 6 months in the community, posi - were group tests 62.17% of control drugs 34.27% of the to compared for tive group intervention ance in the group (control intervention participants it) complete did not through opportunity changes make to as a and volunteering justice, restorative sentence completion following Judge Peer Key fndings Key - statisti were post to pre from Improvements Pre- and post-YSL/CMI changes are only only are changes and post-YSL/CMI Pre- and release Urine analysis—between - self-accept low The Harter revealed profle the Court provided intervention The Teen - ior, honesty, empathy, empathy, honesty, ior, decision- positive making and personal development Children’s own Risk own Children’s Factors and Protective instrument measured: self-image, Positive goal-oriented behav Case Management Case Management (YLS/CMI, Inventory and Andrews Hoge 2002) the citation for provide inventory above Profle (HarterProfle 1985 ) only pre-intervention with the 23 of the group intervention who agreed com - to this. The control plete receive group did not this opportunity Psychological measures measures Psychological utilized Methodist Home for Home for Methodist Youth Level of Service/ Level Youth The article did not Harter Self-Perception - - 0.08) = 0.04) level 0.05) level < = = 0.10) level (analysis (analysis level = 0.10) < 0.001) level VBTE youth (214 days) than the com - (214 days) VBTE youth and this was parison group (182 days) signifcant at the (p 72.8% of high risk in the comparison 57.5% to group compared re-ofended of high risk in the VBTE group VBTE participants, being for reduced for 48% lower to compared 39% lower the group control person-related took even longer, longer, even took person-related signifcant at the (p of re-entry services had re-ofended 49% of the to compared probation- sig - group and this was diference only nifcant at the (p status ofences) excluded (p re-ofend to longer cantly able for 18 participants the from able for group control reofend thanreofend their difer controls; signifcant at the ence statistically (p < 0.06) level their sentences than complete to their signifcant difer statistically controls; ence at the (p Reductions in ofending Reductions Time to re-ofending was greater for greater for was re-ofending to Time Higher risk individuals benefted more; signifcantly re-ofending Violent Those whose previous ofence was ofence was Those whose previous Immediately after mentoringImmediately ended 37% The re-entry group signif - also took - avail only data was NB: Re-ofending Teen Court ofenders less likely to to Court ofenders less likely Teen Court likely ofenders also more Teen - 1 month to 1 year, and it 1 month 1 year, to 4 months of also provides aftercare once the youth returns the to community. participantsVBTE teaches skills through a life ten key includ - system rewards feedback, ing tolerating an “no” for accepting asking permission, answer, instructions, following and conversation developing disagreement skills, ignor ing inappropriate behavior and of others, and respecting helping others (paid mentors) build a (paid mentors) withrelationship the youth in custody are while they and then continue this 6 months mentoring for their release. following During this time, the mentor re-integrative coordinates the to activities relevant supportsand youth - the keep ing of appointments with other agencies decides on the sentence activities) (reparation sentence following volunteer Judge as Peer 3 months Description of intervention Program length from varied Transitional Coordinators Coordinators Transitional Teen Court—A juryTeen of peers including opportunities to sentence length Average - design with a sub- matched - using pro sample pensity modelling pairs match to matched group matched design vention group vention selected the con - group from trol all those referred time non- 1st for ofences felony on all 26 youth on all 26 youth to referred this Teen pilot Court—the Inter Study design Study Quasi-experimental Quasi experimental The study randomly randomly The study The study reportsThe study (SD) 29.2% white, 5.6% Hispanic (SD) 59.45% non-white (SD) fed group; data given not control for group) 15% other Mean age Mean age 14.4 (1.09) 14.4 (0.38) Ethnicity 49.5% black, 15.7% other Mean age 16.32 (1.42) 16.75 (1.32) Ethnicity 40.5% white, Mean age 15.42 (1.45) Age not speci - not Age Ethnicity (Intervention 85% white size: n = 516 size: n = 112 fed = 26 size: n = 52 Participant information M = Male; F = Female Sample size Sample Intervention M/F = 206/52 Control M/F = 212/46 sample Total Sample size Sample Intervention M/F = 45/18 Control M/F = 36/13 sample Total Sample size Sample Intervention M/F = 17/9 Control speci - M/F not Total sample Total (continued) ( 2017 ) Bergseth Bergseth ( 2008 ) Demilio ( 2005 ) 1 Table Author, year, year, Author, location Strom et al. et al. Strom Boufard and Boufard Forgays and Forgays

1 3 Adolescent Research Review Quality score (0– 100%) 58 73 Key fndings Key The study only reports outcomes re-ofending only The study The study only reports outcomes re-ofending only The study - chopathy Checklist: Checklist: chopathy (Forth Version Youth 2003 ) (in press et al. at the time) Checklist: Youth Youth Checklist: (Forth et al. Version 2003 ) of Service inventory Shields and (YO-LSI; 1991 ) Simourd Psychological measures measures Psychological utilized Pre-only: The Hare Psy The Hare Pre-only: Pre-only: Psychopathy Psychopathy Pre-only: Ofender Level Young - 267) following the 267) following = vention group re-ofended vention services) standard (receiving re- ofended intervention; impact on general recidivism in the recidivism on general impact community on violent crime—21% of MJTC to compared re-ofended violently Efect size of violent 49% of controls. ( p = < 0.05) re-ofending Reductions in ofending Reductions 7 out of 10 in the assessed only group 7 out of 10 in the assessed only reportedEfect size not 532 days (SD 532 days 1 in 10 of the group re-ofended DT 2 out of 10 the Mental Health inter 2 years following release there was no there release was following 2 years a clear efect there was However, (DT) aims to break the break aims to (DT) behavior of aggressive cycle in turnwhich should reduce defance and aggression to that thethe individual extent can then access rehabilita - interventions tive the at the interventions Treatment Mendota Juvenile following delivered Centre DT Description of intervention Decompression Treatment Treatment Decompression describe does not The study As above matched group matched - design using pro pensity scoring matched group matched - design using pro pensity scoring Study design Study Quasi-experimental Quasi-experimental - not speci - not fed but used to but used to par match ticipants (SD) however however include as ethnicity in a variable the stepwise logistic regression analysis Mean age/SD Mean age/SD ‘Juvenile’ ‘Juvenile’ ‘Juvenile’ Ethnicity reported, Not Mean age 17.2 (1.0) stated Not Ethnicity reported Not did The study decompression mental health size: n = 30 size: n = 141 Participant information M = Male; F = Female Sample size Sample Intervention Intervention M = 10 Intervention Intervention M = 10 Control M = 10 sample Total Sample size Sample Intervention M = 56 Control M = 85 sample Total (continued) and Van and Van Rybroek ( 2001 ) et al. et al. ( 2006 ) 1 Table Author, year, year, Author, location Caldwell Caldwell Caldwell Caldwell

1 3 Adolescent Research Review Quality score (0– 100%) 59 62 - pro only change; to motivation for control while dropouts improved gram completers theirincreased ofending Key fndings Key The study only reports outcomes re-ofending only The study The study only reports outcomes re-ofending only The study been useful to have it would suggest Authors Screening Inventory– Screening Child / Self-Report (TESI–C / SR; Ford 1999) et al. the citation for provide inventory above Second Version et al. Ford (MAYSI–2; 2008 ) Psychological measures measures Psychological utilized Pre-only: Events Traumatic does not The study Youth Massachusetts Instrument,Screening None - 0.05) level. How 0.05) level. = into theinto community), re-ofending the however ‘declined signifcantly’ no further provides details study but only for program completers; program completers; for but only those for rate reconviction one-year lower signifcantly was completing comparisonthan the group, matched signifcant at the (p ever, after 2 years these diferences after 2 years ever, exist had ceased to dropouts after excluding ments even Reductions in ofending Reductions 6 months later those released (data included only Only ETS showed reduced ofending, reduced showed ETS Only - no signifcant improve R&R showed - - processing and strengths- processing based reintegration, memory decrease to re-examination rumination, panic or dis - sociation sessions can several pleted new for act as peer coaches participants (ETS)—20–22 sessions. (ETS)—20–22 teaching intervention Group thinking patterns and cogni - skills through impulse tive thinking, fexible control, reasoning, and moral values interpersonal solv problem tak ing, social perspective ing, and critical reasoning (R&R)—36 sessions. Group how teaching intervention using emotions address to of risky instead reasoning or violent behavior Description of intervention Self-regulation, trauma trauma Self-regulation, who had com - Those youth Enhanced Thinking Skills & Rehabilitation Reasoning matched group matched design design Study design Study Quasi experimental Matched group Matched (SD) fed Ameri - can 32% Hispanic, 24% white add (fgures 99) up to discharge (SD) nicity was nicity was included as in a variable the logistic regression analysis Mean age Mean age 14.4 (.98) Age not speci - not Age Ethnicity 43% African at Mean age 20.40 (1.6) 20.50 (1.79) Ethnicity reported Not - eth However, fed = 197 size: n = 394 size: n = 3068 Participant information M = Male; F = Female Sample size Sample Intervention M/F = 179/18 Control speci - M/F not Total sample Total Sample size Sample Intervention M = 1534 Control M = 1534 sample Total (continued) Hawke Hawke ( 2012 ) ( 2005 ) 1 Table Author, year, year, Author, location Ford and Ford Cann et al. Cann et al.

1 3 Adolescent Research Review Quality score (0– 100%) 60 - useful to use recognized measures of inter measures use recognized useful to as the positive change, to nal motivation - moti pre-existing be due to outcomes may rather than the program change to vation Key fndings Key The study only reports outcomes re-ofending only The study been have The authors conclude that it would Psychological measures measures Psychological utilized None days to re-arrest were 278 for the 278 for re-arrest to were days class plus cell phone group, 191 for the group, and 106 for the class-only group control phone calls had signifcantly receive group the (45%) but due to small not size, thissample was diference signifcant statistically half of at least group who answered re-arrested, this calls were their daily signifcant at the a statistically was (p = 0.83) level less than half group who answered re-arrestedtheir phone calls were re-arrested were arrested Reductions in ofending Reductions 1 year following the mean intervention, following 1 year Those who attended the class but did not total rearrests than the control fewer 39% of those in the class plus phone 80% of those in the class plus phone 55% of those the receiving class-only re- 90% of the group were control ral training sessions aiming training ral individualized create to long-term goals group received phone-coach twice-daily automated a year phone calls for Description of intervention 6 × 90 min cognitive - behavio be in the to Those chosen groups the study used random assignment, the next then for three identifed controls matched intervention group, the gave researchers 28. a phone to The remaining 11 made up the group ‘class-only’ Study design Study For the frst two thetwo For frst Of the 39 in the - (SD not (SD not provided) taken from from taken the initial report (Bur et al. raston 2012 ) Mean age Mean age 16.1 16.5 15.7 Mean ages Mean ages Ethnicity reported Not class plus of phone 1 year calls class only size: n = 70 Participant information M = Male; F = Female Sample size Sample Intervention— M/F = 20/8 Intervention— M/F = 10/1 Control M/F = 27/4 sample Total (continued) et al. et al. ( 2014 ) 1 Table Author, year, year, Author, location Burraston

1 3 Adolescent Research Review Quality score (0– 100%) 68 measure motivation and determine whether motivation measure is associated with involvement phone-coach the and whether association motivation, and rearrests is calls answered between compared individuals with low for diferent with high motivation Key fndings Key Authors concluded it would be useful to be useful to concluded it would Authors Psychological measures measures Psychological utilized None < 0.01) level pared to the group to pared control additional 100 calls answered every of the reduced likelihood signifcantly 36%, and this arrest was by a felony signifcant at the (p Reductions in ofending Reductions Overall ofending did not reduce com - reduce ofending did not Overall in those phone calls, receiving However, Description of intervention As above for all participants for apart 20 from the assigned by court specifcally theto - interven tion condition Study design Study Random assignment Random 45% non- white Mean age/SD 16.07 (1.33) 16.65 (1.28) 17.67 (1.09) Ethnicity 55% white, report on the demographics of three sepa - groupsrate that intervention and control participants came from: grams = 57 was 270, was the authors 14 excluded incom - due to dataplete but it specifed is not which from group) ple size: Participant information M = Male; F = Female Sample size: Sample The authors = 73 Probation - pro Rural Secure Secure = 126 care (original sample Total fnal sam - Total n = 256 (continued) ( 2015 ) 1 Table Author, year, year, Author, location Bahr et al. Bahr et al.

1 3 Adolescent Research Review regards to control group selection, twelve studies used a regarding gender totaled 36 participants. One study was not matched control, one used random allocation (Bahr et al. able to specify the gender of 10 participants (Haines et al. 2015) and one randomly selected the control group (Forgays 2015). One study did not report control group gender (For- and DeMilio 2005). However only six of those studies using gays and DeMilio 2005). matched controls controlled for baseline diferences (Cald- Across all studies, age ranged from 14 to 21 years old. In well and Rybroek 2001; Caldwell et al. 2006; Farrington the eight studies that reported age of both intervention and et al. 2002; Haines et al. 2015; Hoogsteder et al. 2018; Strom control groups, the average age of the intervention group was et al. 2017). 16.17 and the average age of the control group was 16.72. Finally, one study did not report age at all, simply stating Settings they were “juvenile” (Ford et al. 2008). With regards to ofending histories, eleven participant Of the studies identifed, fve were based in the community groups met the criteria for persistent or repeat ofenders (Boufard and Bergseth 2008; Burraston et al. 2014; Forgays meaning they had ofended at least once previously (but and DeMilio 2005; Haines et al. 2015; Hubble et al. 2015). usually more than once). Participants from one group were However, one of those studies contained youth nearing the in the “early stages” of delinquency (Lindblom et al. 2017). end of their custodial sentence before continuing in the com- One intervention did not include details of previous con- munity (Boufard and Bergseth 2008). Participants in seven victions but did take place in a young ofender institution of the studies resided in custodial placements. That is, three suggesting the youth were either serious ofenders, or had studies used participants from standard “young ofender ofended more than once (Cann et al. 2005). One interven- institutions” (Cann et al. 2005; Farrington et al. 2002; tion described the ofences as “relatively minor but leading Hoogsteder et al. 2018), one examined participants from to a reprimand or fnal warning from the police” (Haines placements described as “residential homes” (Strom et al. et al. 2015, p. 127). Participation appeared to be voluntary 2017), and three consisted of therapeutic custodial place- for seven of the intervention groups (Bahr et al. 2015; Bur- ments (Caldwell and Rybroek 2001; Caldwell et al. 2006; raston et al. 2014; Cann et al. 2005; Farrington et al. 2002; Ford and Hawke 2012). The two remaining studies recruited Ford and Hawke 2012; Haines et al. 2015; Lindblom et al. youth from both the community and custodial placements 2017). For fve, it appeared to be compulsory (Caldwell and (Bahr et al. 2015; Lindblom et al. 2017). Rybroek 2001; Caldwell et al. 2006; Forgays and DeMilio 2005; Hoogsteder et al. 2018; Strom et al. 2017). It is unclear Country of Origin whether the intervention was voluntary or compulsory for the remaining two groups (Boufard and Bergseth 2008; Of the interventions, eight originated from the USA (Bahr Hubble et al. 2015). et al. 2015; Boufard and Bergseth 2008; Burraston et al. 2014; Caldwell and Rybroek 2001; Caldwell et al. 2006; Race Ford and Hawke 2012; Forgays and DeMilio 2005; Strom et al. 2017), four from the UK (Cann et al. 2005; Farrington One study reported that 86.6% were Scandinavian, 6.7% et al. 2002; Haines et al. 2015; Hubble et al. 2015), one from were East European and 6.7% were African (Lindblom the Netherlands (Hoogsteder et al. 2018), and one originated et al. 2017). Another study reported that 29.5% were Dutch, from Sweden (Lindblom et al. 2017). 23.8% were Moroccan, 15.2% were Surinam, 6.7% were Turkish, 6.7% were Dutch Antillean, 2.9% “other: western” Population and 15.3% “other: nonwestern” (Hoogsteder et al. 2018). Of the 8 studies conducted in the U.S.A., 5 report race. The number of intervention participants totaled 2837, and That is, one recorded their sample as being 49.5% black, the number of control participants totaled 2715 making a 29.2% white, 5.6% Hispanic and 15.7% “other” (Strom et al. total of 5552. Of those detailing gender, fve of the studies 2017). One study noted that their sample contained 43% contained male participants only totaling 3582 (Caldwell African American, 32% Hispanic and 24% white (Ford and and Rybroek 2001; Caldwell et al. 2006; Cann et al. 2005; Hawke 2012). Other studies contain less detail; one stated Farrington et al. 2002; Hubble et al. 2015) and nine included that participants were 55% white and 45% non-white (Bahr both genders of which 1587 were male and 337 female (Bahr et al. 2015). Another study reported race as 40.5% white and et al. 2015; Boufard and Bergseth 2008; Burraston et al. 59.45% non-white (Boufard and Bergseth 2008). Finally, 2014; Ford and Hawke 2012; Forgays and DeMilio 2005; one study recorded the race of their intervention group as Haines et al. 2015; Hoogsteder et al. 2018; Lindblom et al. 85% white and 15% “other” (Forgays and DeMilio 2005). 2017; Strom et al. 2017). Across all reported studies, there Although they used race to match intervention and con- were 5169 male participants and 337 females. Missing data trol groups, three studies did not include this data in their

1 3 Adolescent Research Review articles (Burraston et al. 2014; Caldwell and Rybroek 2001; 14% (220) of program starters dropped out (subsequently Caldwell et al. 2006). Of the four studies from the U.K, two excluded from the analysis) (Cann et al. 2005). One study include race data. One study recorded participants race as reported that out of the 176 who started the program, 71 did 87.7% white and 12.3% non-white (Farrington et al. 2002). not complete (Farrington et al. 2002). Although this gives The second study recorded participants as 69.6% white, an attrition rate of 40.34%, the decision to include this study 20.1% non-white/“mixed” and 10.3% as “other/other Euro- relates to its valuable psychological pre- and post-measures. pean” (Haines et al. 2015). Although it used race data as a Another study reported that 26 out of the 27 repeat ofend- variable for regression analyses, one study did not report ers appeared for their Teen Court sentence and their fnal participants race, or whether this afected intervention out- analysis consisted of 24 (81%) participants who completed comes (Cann et al. 2005). Finally, one study did not specify their sentences (Forgays and DeMilio 2005). The fnal study race in relation to either participant matching or intervention to report attrition stated that no participants dropped out outcomes (Hubble et al. 2015). Overall, in the USA, studies (Hubble et al. 2015). reported 46% participants as white and 53.06% as non-white, while in the U.K. these fgures were 78.65% and 21.35% Intervention Type and Efcacy respectively. However, the U.K. fgures do not necessarily give a true picture as overall in the UK, 40% of young peo- The types of intervention employed largely addressed areas ple in custody are from black, Asian and minority ethnic such as cognitive abilities, mentoring and to a lesser extent, (BAME) backgrounds (Ministry of Justice 2016). restorative justice. Interventions therefore varied greatly in content, sample size, location and measurement of efcacy. Pre‑ and Post‑psychological Measures To examine the efectiveness of the interventions reported, the results section below frst discusses studies that uti- Five of the research studies identifed reported pre- and post- lize psychological assessments pre- and post-intervention, psychological assessment for both intervention and control followed by those using pre-assessment only. Finally, the groups (Farrington et al. 2002; Haines et al. 2015; Hoogst- results section discusses interventions using no pre- or post- eder et al. 2018; Hubble et al. 2015; Lindblom et al. 2017). assessments. Although it may appear that reporting on such A further two studies reported pre- and post-psychological studies may be counter-intuitive, they do fall under the cat- assessment for the intervention groups only (Boufard and egory of studies which show promise. Bergseth 2008; Strom et al. 2017). A further four studies reported pre-intervention psychological assessment but not Studies Reporting Pre‑ and Post‑psychological post-assessment (Caldwell and Rybroek 2001; Caldwell Measures on Both Intervention and Control Groups et al. 2006; Ford and Hawke 2012; Forgays and DeMilio 2005). The remaining three studies did not include any One intervention consisted of using a facial emotion rec- measures other than re-ofending, but nevertheless targeted ognition tool in the form of computerized slides to assess predominantly psychological factors in the intervention. As emotional recognition in others (Hubble et al. 2015). At pre- these studies showed a reduction in re-ofending and focused test, results revealed both groups (intervention and control) specifcally on cognitive and emotional well-being, these were poor at recognizing fear, sadness and anger in others. studies met the criteria for inclusion (Bahr et al. 2015; Bur- A 2-week course then trained participants to recognize facial raston et al. 2014; Cann et al. 2005). expressions. At post-test, the intervention group signifcantly improved their ability to recognize facial expressions of fear, Attrition anger and sadness, while their controls either remained the same (in relation to fear and anger) or worsened (in relation Seven studies reported on participant dropout (Bahr et al. to sadness). Results showed the volume of re-ofending in 2015; Burraston et al. 2014; Caldwell and Rybroek 2001; both groups signifcantly decreased 6-months post-interven- Cann et al. 2005; Farrington et al. 2002; Forgays and tion. However, ofences committed by the intervention group DeMilio 2005; Hubble et al. 2015). One study reported that decreased signifcantly in severity. 14 participants began but did not complete the study, either Another study to utilize a pre- and post-measure design because they moved out of state of decided not to continue tested the efcacy of a 25-week High Intensity Training (excluded from the fnal analysis) (Bahr et al. 2015). Another (HIT) intervention (Farrington et al. 2002). In addition to study reported that just 2 control and 4 intervention partici- daily military-style training, this intervention also consisted pants dropped out from their study (i.e., moved away from of Enhanced Thinking Skills in conjunction with a 1-week the area) leaving a fnal sample of 70 (Burraston et al. 2014). outward-bound style camping expedition that culminated in One study reported that no participants left the interven- a work placement. Outward bound interventions may help tion group (Caldwell et al. 2006). Another reported that youth who have experienced difculties by providing an

1 3 Adolescent Research Review environment that contains a sense of stability which fosters as motivation, aggression, self-control and group work, plus positive interactions between staf and youth (Trundle and optional modules which they could choose from (reducing Hutchinson 2020). stress, controlling impulses, re-interpretation of events, regu- The assessments conducted on both intervention and con- lating emotions, confict management and a systemic family trol groups included emotion control, adaptation (assessing module). The intervention also included a focus on drama how well both groups adjusted to incarceration), and crimi- and mindfulness classes. Pre- and post-measures revealed nal thinking styles. Results revealed that the intervention signifcant reductions in risk, suggesting more pro-social group reported better control of aggression and being less coping and stress management was a contributing variable. anti-staf (indicating greater levels of adaptation) than the After 1 year there were no signifcant reductions in recidi- control group. In concert, these fndings suggested a more vism; improvements only became apparent 2 years following positive adjustment to the custodial and the HIT regime for the custodial placement. At this point, the authors noted a the intervention group. The authors also noted that there signifcant reduction in general recidivism in the Re-ART were increases for the intervention group in impulsivity, jus- group compared to the control group (p < 0.001), whereby tifcation of crime, and belief society owed them a living. 82.1% of the control group had reofended with a general These changes were signifcant at (p = 0.085) and (p = 0.008) ofense compared with 44.4% of the Re-ART group. In respectively. Even so, following the HIT regime, predicted the same time period, there was also signifcant reduction re-ofending compared with actual re-convictions showed in violent recidivism for the Re-ART group compared to that re-convictions were signifcantly lower for the interven- the control group (p < 0.05). After 3 years, both reductions tion group compared to the control group 12 months later. remained signifcant (p < 0.05). The groups did not difer on There were no diferences in ofending outcomes for violent/ recidivism regarding property crimes with violence after 1, non-violent ofenders. However, 24-months post-interven- 2 or 3 years (Hoogsteder et al. 2018). tion these diferences were no longer statistically signifcant The fnal study to use a pre- and post-test matched group in either group (i.e., both groups re-ofended as predicted) design examined the benefts of a mental health diversion (Farrington et al. 2002). intervention (Haines et al. 2015). This intervention involved The third study to use a pre- and post-measures matched targeting children and youth with mental health and/or group design tested the efcacy of “A New Direction” (Lind- developmental problems as soon as they entered the youth blom et al. 2017). Psychological aspects of this intervention justice system. Youth practitioners referred participants to included promoting life ambitions, challenging criminal one of four Youth Ofending Teams engaged in this study. ideas, cognitive behavioral therapy, increasing empathy Participants then received improved access to specialized (efects on the victims of crime) and problem solving. In services such as the Child and Adolescent Mental Health addition, it addressed advantages and disadvantages of Services (CAMHS) and/or referral to other relevant statutory crime, communication with family and friends, and gen- or voluntary agencies. The authors regressed data gathered eral social skills training over 9–30 weeks. In order to from the Youth Ofending Teams’ own database pertaining assess psychological changes, the authors examined Sense to “areas of concern and vulnerability” in relation to both of Coherence, which assesses whether life makes sense, is mental health and other risk areas against re-ofending for manageable, and has meaning (Eriksson 2016). The authors the entire sample. Results revealed that the only variable that also assessed criminal thinking styles (Walters 2002). Com- signifcantly positively correlated with continued ofending pared to their controls, the intervention group experienced was previous ofending. However, mental health factors, signifcant reductions in criminal thinking and a signifcant specifcally being unhappy, dissatisfed, and having low increase in Sense of Coherence. The authors report no sig- self-esteem had a positive relationship with re-ofending. nifcant changes in the control group. Twelve months later, Following the mental health diversion intervention, analysis the intervention group continued to show sustained desist- of re-ofending data 15–30 months later took place. Despite ance in that just one out of the eleven participants had re- re-ofending rates remaining equal in both intervention and ofended. By 24 months, out of the 8 in the intervention control groups, the intervention group took signifcantly group for whom data were available, none had re-ofended. longer to re-ofend than their controls in two of the Youth In the control group, during the 12 months following the Ofending Teams (580 days vs 334 days and 220 days vs intervention, 50% had re-ofended and at a higher rate (Lind- 84 days) (Haines et al. 2015). blom et al. 2017). A study evaluating an intervention called Responsive Studies Reporting Pre‑ and Post‑psychological Aggression Regulation Therapy (Re-ART) which specif- Assessment for Intervention Groups Only cally targets aggressive behavior, included pre and post- measures of violence risk to assess its efcacy (Hoogsteder A study assessing the impact of Value-based Therapeutic et al. 2018). Participants received core training in areas such Environments (VBTE) on re-ofending employed a pre- and

1 3 Adolescent Research Review post-test matched group design (Strom et al. 2017). VBTE recidivism with repeat offenders (Forgays and DeMilio is a multi-purpose hybrid behavioral model incorporating 2005). The aim was reintegrating the individual into society both value-based and skills-based CBT components via an through reparation activities, described as being “socially individualized approach. VBTE teaches participants ten key and personally” challenging for the youth, but ultimately life skills through a rewards system that includes tolerating empowering (e.g., writing letters of apology). A unique fac- feedback, accepting “no” for an answer, asking permission, tor in this Teen Court is that following successful completion following instructions, developing conversation and disagree- of their sentence, participants can then become peer judges ment skills, ignoring inappropriate behavior of others, and themselves. Teen Court Jurors may therefore include former respecting and helping others. Program length varied from youth who have ofended. In this study, self-worth profles one month to one year, and the youth received four months assessed prior in the intervention group showed that these of aftercare on their return to the community. In terms of psy- youth had low self-acceptance (suggesting they would like chological changes, the authors report signifcant improve- to be diferent from how they were) (Harter 1985). Follow- ments in self-image, goal orientation, honesty, empathy, posi- ing their sentencing, a signifcant and unexpected number tive decision-making, and personal development. The largest of participants chose to continue their involvement as peer improvement was in positive decision-making which increased jurors. Re-ofending results showed that just 12% of the from 4 to 43%. In terms of re-ofending, results revealed that intervention group had re-ofended 6-months later (mostly the number of days until the frst re-ofence was signifcantly theft related). In the control group, 38% re-ofended with greater for the VBTE group (214 days) than the control group theft or assault. The authors surmise that low self-worth, (182 days). Further, 57% of high risk/need youth in the VBTE coupled with a desire to change, ofers a possible psycho- group re-ofended compared to 73% of high risk/need youth logical explanation for the efectiveness of this Teen Court in the control group. The program also reduced the likelihood (Forgays and DeMilio 2005). This led the authors to suggest of a violent re-ofence charge by 67% and any new charge by that the role of personal empowerment may help enable anti- 49%. In other words, those youth whose risks and needs were social youth to adopt prosocial values via accessible pro- greater beneftted more from the program (Strom et al. 2017). social opportunities (cf. Mohajer and Earnest 2009). An evaluation of the impact of a “Re-entry” intervention A further study to employ a matched group design tested also used a pre- and post-test matched group design with a the efcacy of Decompression Treatment (DT) on re-ofend- control and an intervention group to test its impact on re- ing (Caldwell and Rybroek 2001). This design included two ofending (Boufard and Bergseth 2008). This intervention control groups whereby one received standard therapeutic involves transitional coordinators (paid mentors) building intervention services in the same juvenile center as the a relationship with the youth while they are in custody and intervention group, while the second was based in a con- then continuing this mentoring for 6 months following their ventional correctional facility. DT focuses specifcally on release. During this time, the mentor coordinates re-integra- reducing defance and targets the hardest to manage youth tive community activities relevant to the young person. In who are unresponsive to standard therapeutic intervention. terms of psychological change, there was a signifcant reduc- The program developers theorize that punishment increases tion in risk/need scores of 17% for the intervention group. anti-social behavior and serves to further increase antago- However, the study did not state which specifc social or psy- nism towards society. Decompression therefore refers to the chological risks/needs reduced and as the control group did method of breaking this cycle of aggressive behavior (Cald- not receive the same assessment, we cannot say that these well and Rybroek 2001). This in turn should reduce defance changes were due to the intervention. In terms of re-ofend- and aggression to the extent that the individual can then ing, results following the re-entry intervention showed that access rehabilitative interventions. As well as a behavioral during the 6 months following release from custody, a statis- outcome, a reduction in defance also suggests an improved tically signifcant between group diference occurred where psychological state in the context of a custodial placement. 37% of the intervention group re-ofended compared to 49% Re-ofending data showed that after 532 days post-interven- of the control group. However, as there are no re-ofending tion, 10% of the DT intervention group re-ofended, 20% results following the cessation of the mentoring support it is of those receiving standard therapeutic intervention had re- unclear if changes sustained beyond the 6 months (Boufard ofended, and 70% of the control group receiving standard and Bergseth 2008). correctional rehabilitation had re-ofended (Caldwell and Rybroek 2001). Studies Reporting Psychological Measures A further evaluation of DT with a larger sample matched Pre‑intervention Only intervention and control groups on predictions of re-ofend- ing, IQ levels, conduct disorder symptoms and substance A pilot evaluation of “Teen Court”, normally utilized with abuse (Caldwell et al. 2006). Both groups indicated high first time offenders, demonstrated success in reducing probability of re-ofending, below average IQ levels, high

1 3 Adolescent Research Review conduct disorder symptoms and extremely high levels of implement psychological interventions that showed reduced substance abuse. Results 2 years following DT showed that re-ofending. 57% of the intervention group re-ofended in the institution An evaluation using a matched group design to evalu- or community compared with 78% of their controls. This ate two CBT interventions, whereby one group received between group diference was signifcant at the (p < 0.01) Enhanced Thinking Skills (ETS) and the other received level. However, following the use of propensity score match- Reasoning and Rehabilitation (R&R), showed that only ETS ing, the intervention showed no impact on general recidi- reduced recidivism (Cann et al. 2005). Participants receiving vism in the community. Nevertheless, the efects on violent ETS took part in a group-based program for 20 × 2-h ses- re-ofending remained signifcant (see Table 1). Specifcally, sions where they learnt thinking patterns and cognitive skills 2 years after release, 10% of the control group accounted for through impulse control, fexible thinking, values and moral 16 homicides while none of the intervention group received reasoning, interpersonal problem solving, social perspec- homicide charges (Caldwell et al. 2006). Although both tive taking, and critical reasoning. R&R taught participants studies draw attention to the relevance of reduced defance, to address their emotions using reasoning instead of risky they did not specify the therapeutic intervention received or violent behavior over 36 × 2-h sessions. Re-ofending following DT, nor conduct post-measures, hence it is dif- results 12 months later showed that there was no signifcant cult to pinpoint distinct psychological factors following the reduction in re-ofending in the group who received R&R intervention that led to the reduction in criminal activity. compared to their matched comparisons, even after exclud- A fnal study to utilize psychological assessment pre- ing program dropouts. Of those who completed ETS (i.e., intervention only also utilized a matched-group design to excluding program dropouts), 31.4% re-ofended compared assess the impact of Trauma Afect Regulation Guide to to their controls of whom 35.5% re-ofended (between group Education and Training (TARGET) on re-ofending (Ford efect was signifcant at the p < 0.05 level). Conversely, pro- and Hawke 2012). TARGET in youth justice settings fol- gram dropouts across both samples increased their ofending lows the assumption that problems causing youth to engage by 47%. However, 2 years later the positive efects noted in delinquent behavior are largely a result of unrecog- in the ETS group had disappeared leading the authors to nized stress reactions (Andershed et al. 2008). Participants conclude that further refresher sessions may be necessary received up to 10 TARGET sessions which included train- to increase the likelihood of sustained change (Cann et al. ing in self-regulation, trauma processing and strengths-based 2005). reintegration. To increase sense of control, awareness and An initial evaluation of the efcacy of a CBT program safety, TARGET also includes memory re-examination called RealVictory, showed that engagement in 6 × 90 min procedures which aim to decrease rumination (dwelling on cognitive behavioral training sessions followed by daily past upsets), panic or dissociation. To act as role models, automated phone calls for a year, reduced subsequent arrests all staf members (including caretakers and administration by 51% (Burraston et al. 2012). In this study, 39 youth staf) underwent TARGET training by learning and practic- received the RealVictory program which included personal ing the techniques along with the youth. In addition, those support in creating individualized long-term goals. Follow- youth who had completed several sessions can act as peer ing this, 28 received cell phones and received a twice daily coaches for new participants. The authors report that re- phone call regarding goal progress for one year. The remain- ofending declined signifcantly following implementation ing 11 formed a “class-only” intervention group who did not of TARGET, but did not provide specifc data. In addition, receive the phone option. The control group contained 31 the study only includes recidivism data for those youth who juveniles who received standard probation intervention (Bur- returned to the community, excluding those who moved to raston et al. 2012). A further interpretation of the impact of another detention center. After controlling for diferences the phone calls involved dividing those who received phone between the groups, results revealed that participating in calls into two groups; “high” (answering over half their a single session of TARGET in the frst 14 days of deten- phone calls) and “low” (answering less than half of their tion was associated with 0.53 fewer disciplinary incidents phone calls). Results showed that after 12 months, those who and 69 fewer minutes of disciplinary seclusion (Ford and answered more than half their daily calls had the lowest re- Hawke 2012). arrest rate of 39%. Those who answered less than half their calls had a re-arrest rate of 80%, a similar rate to the control group receiving standard probation who had a re-arrest rate Studies Measuring Re‑ofending Only of 90%. The group who participated in the intervention only (i.e., did not receive a cell phone) had a re-arrest rate of 55% The following studies did not use psychological assessment (Burraston et al. 2014). to record pre- or post-intervention changes, but they did A replication of the RealVictory intervention employed an RCD approach with a much larger sample (Bahr et al.

1 3 Adolescent Research Review

2015). However, results showed that after 12 months, the One study reported that increasing the ability to recog- group receiving ‘RealVictory’ plus phone calls did not nize facial expressions (fear and anger) led to reduced sever- reduce their re-ofending compared to the control group. ity of re-ofending (Hubble et al. 2015). Related research To determine whether the number of calls answered had provides further evidence for the benefts of emotion skills any relationship with re-ofending, the authors undertook training for youth with callous-unemotional (CU) traits further analysis. This analysis revealed the total number of (Lui et al. 2019). Specifcally, intervention participants sig- calls answered by the RealVictory group, although it did not nifcantly increased perspective-taking compared to their achieve statistical signifcance in relation to general recidi- controls. Further, the control group reported lower levels vism, it did signifcantly reduce felony arrests. Specifcally, of self-reported empathy and pro-social behavior, whereas for every additional 100 calls answered, the likelihood of the intervention group either declined less or had minimal a felony arrest signifcantly reduced by (Bahr et al. 2015). change (Lui et al. 2019). These results refect those reported in in the results section, whereby emotional skills training appeared to have prevented deteriorating symptoms of CU Discussion (Hubble et al. 2015). Programs to augment the emotional development of youth may result in increased likelihood of Research has shown that experiencing psychological trauma empathizing with and behaving pro-socially towards oth- at a young age can hinder normative adolescent psychologi- ers thus decreasing the likelihood of antagonistic behaviors cal development, leading to low levels of self-concept and (Lui et al. 2019). Relatedly, the Decompression Treatment emotional regulation (Gibson and Clarbour 2017). These intervention also found targeting of defance plus therapeutic factors have been associated with youth who ofend and intervention reduced the severity of re-ofending (Caldwell may be a pre-cursor to delinquent behavior such as avoidant et al. 2006). coping (e.g., taking drugs) or aggression (Carr et al. 2001). In contrast to the New Directions intervention, the HIT Hence, the purpose of this article was to systematically intervention found certain aspects of criminal thinking such review relevant interventions targeting the development of as justifcation of crime did not reduce (Farrington et al. specifc psychological characteristics hypothesized to reduce 2002). However, it did provide further evidence that psycho- re-ofending. A total of 14 studies containing 12 diferent logical improvements (e.g., being less anti-staf and having interventions met the criteria for inclusion. These studies better control of aggression) related to reduced re-ofending. showed that re-ofending reduced in interventions that dem- In line with strain theory, increased pro-social behaviors may onstrated signifcant increases in psychological resources. enable youth to achieve goals via agreeable ways such as However, it was not always easy to identify which areas of through pro-social relationships (Boeck et al. 2008). For psychological development were associated with reduced example, the evaluators of the Re-entry intervention sug- re-ofending (e.g., Bahr et al. 2015; Boufard and Bergseth gest its success may have been due to a specifc focus on 2008; Burraston et al. 2014; Caldwell and Rybroek 2001; facilitating community networking (Boufard and Bergseth Caldwell et al. 2006; Cann et al. 2005; Ford and Hawke 2008). Similar intensive aftercare programs that were not 2012; Forgays and DeMilio 2005; Haines et al. 2015). successful in reducing re-ofending did not implement the Other studies were much clearer in demonstrating this creation of community support networks in any systemic link. For example, the program “A New Direction” increased way (Wiebush et al. 2005). sense of coherence (SOC) and reduced criminal thinking Both the Re-ART and VBTE interventions clearly dem- styles leading to near total desistance up to 2-years post- onstrated that improvements in negative attitudes, personal intervention (Lindblom et al. 2017). A recent systematic development and positive decision making related to a sig- review found that youth who were depressed or anxious, nifcant reduction in ofending. Specifcally, Re-ART suc- misused drugs and alcohol, had poor social skills or con- ceeded in reducing violent and general re-ofending across duct problems, or engaged in delinquent behavior were also the sample (Hoogsteder et al. 2018). VBTE succeeded in more likely to have low sense of coherence (Lansimies et al. reducing re-ofending only in participants with a higher 2017). Despite SOC developing during adolescence, studies risk of re-ofending and reported the largest improvement have reported that SOC (and reductions in re-ofending) can was in positive decision-making (Strom et al. 2017). Re- increase following intervention with adult men (Lindblom ART’s program evaluators suggest its success may be due its et al. 2018). Furthermore, in a study of incarcerated women, responsivity-focused approach, which connects to the ado- increased SOC related to a signifcant decrease in global lescent’s frame of reference (Hoogsteder et al. 2012, 2018). emotional distress (Höjdahl et al. 2015). Therefore, SOC This approach involves staf seeing the youth’s perceptual appears to have strong grounding for subsequent intervention world as they do, facilitating afnity with their frame of work in these at-risk populations. reference (e.g. Bowen et al. 2013a, b). The cultural training

1 3 Adolescent Research Review provided also enabled program staf to include ethnic sensi- re-ofence (e.g., Haines et al. 2015). Several researchers tivity into their approach. implied intervention success by noting reductions in the The TARGET intervention, which focused on psycho- severity of the re-ofence or separated violent from non- logical well-being and included the opportunity for youth to violent re-ofending (e.g., Caldwell et al. 2006). All stud- become peer mentors following participation, reduced sub- ies except one used ofcial arrests or court data that mostly sequent disciplinary problems in those participating (Ford consisted of police records. However, ofence reporting and Hawke 2012). A separate study evaluating TARGET in systems may mean infated intervention success due to the a youth detention center reported both decreased negative length of time between the actual ofence being commit- afect (e.g., depression, defance, aggression) and increased ted and the criminal charge made (St. James-Roberts et al. positive afect (e.g., optimism) for the intervention group 2005). Going forward, it is likely that studies should report only (Marrow et al. 2012). A further strength of TARGET intervention success by several outcomes (e.g., time and is that all staf members and those youth who had completed severity) using both ofcial and self-report. several sessions can act as peer coaches for new participants, Fundamentally, no interventions assessed long-term thereby increasing the support networks available. psychological changes post-intervention. In other words, A potential moderating variable applicable to all inter- it is difcult to know how long intervention efects lasted. vention success is the individual’s motivation to engage Therefore, it is not possible to ascertain whether re-ofend- (Bahr et al. 2015; Burraston et al. 2014; Cann et al. 2005). ing was due to the loss of psychological skills developed, One way to increase motivation to change is to increase or inefective intervention designs. Previous studies have empowerment, via for example perceived power and con- argued that using only recidivism to measure success is trol (Mohajer and Earnest 2009). The Teen Court reported of limited use, and that identifcation of factors such as on in this review ofered participants the opportunity to psychological resilience related to desistance may improve become peer judges following sentence completion, sug- both measurement of success, and understanding of desist- gesting increased perceptions of empowerment may have ance (Farrington et al. 2000). led to the higher than expected number of youths taking this opportunity (Forgays and DeMilio 2005). Empowerment is an additional theoretical perspective to those normally Additional Support Post Intervention Increases cited for Teen Courts (peer/procedural justice, deterrence, Success labeling, restorative justice, law-related education, and skill building; Butts et al. 2002). One other potential moderating factor to successful inter- This review has highlighted the importance of matching vention outcome is the inclusion of additional support interventions to individual needs. For example, intensive post-intervention. For example, participants in the VBTE aftercare programs seem more efective when they include study benefted from four months of aftercare after return- community integration and R&R has greater success with ing to the community (Strom et al. 2017). The Re-entry adults. Likewise, Teen Courts appear more efective with intervention utilized paid mentors (Boufard and Bergseth repeat ofenders rather than frst-time ofenders (who nor- 2008). Evaluation of RealVictory found that individual tel- mally receive this intervention). A meta-analysis of Teen ephone calls reinforcing goal setting reduced re-ofending Court reported non-signifcant treatment efects, but in look- (for those motivated to answer) (Bahr et al. 2015; Burras- ing specifcally at those including repeat ofenders, efects ton et al. 2014). In line with strain theory, youth returning become signifcant (Bouchard et al. 2017). This draws atten- to communities where opportunities for crime thrive may tion to the relevance of matching intervention content, style, require greater support against anti-social infuences than and intensity to each youth on a case by case basis and may those experiencing fewer negative infuences (Agnew et al. help to explain discrepant fndings in intervention studies. 2001). Several intervention evaluators noted this issue (Caldwell et al. 2006; Hoogsteder et al. 2018; Strom et al. No Consensus in Reporting Recidivism/Little 2017). Adolescence is a time of exceptional psychologi- Consideration of Psychological Changes cal development, and further scafolding and support via environments that bolster opportunities to thrive may be Of the studies reviewed, follow-up data regarding re- fundamental to sustained change (National Academies of ofending post-intervention varied from 6 to 36 months. Sciences, Engineering, & Medicine 2019). Future studies Intervention success (re-ofences recorded) and the def- should therefore conduct follow-up support and psycho- nition of assessment of intervention success also varied logical assessment, along with re-ofending data at 3–6 from study to study. That is, some recorded the number of monthly intervals over a prolonged period (3 years is the re-ofences over a set time period (e.g., Farrington et al. maximum follow-up period noted in this review). For 2002). Some reported the number of days before the frst example, a reassessment of the long-term benefts of the

1 3 Adolescent Research Review

HIT intervention 10 years post-intervention found inter- Need for Strong Evaluations of Interventions vention efects in reducing reofending diminished after for Females 4 years (Jollife et al. 2013). Another important fnding was that most participants in the Interventions Successful in Reducing Severity studies reviewed were male. Specifcally, out of the studies of Re‑ofending that reported gender, 93% were male and just 7% female, meaning that fndings in this review may not generalize This review has also shown that some interventions are to females. Evidence on whether interventions need to be successful in reducing the severity of the criminal activity gender specifc is mixed. For example, some studies show itself. For example, several studies found that while levels male and female adolescents have diferent risks and needs of re-ofending did not change, violent or serious recidivism (cf. Vitopoulos et al. 2012). Others fnd that although male reduced signifcantly following intervention (Bahr et al. and female youth have difering protective factors related to 2015; Caldwell et al. 2006; Hubble et al. 2015; Strom et al. desistance (e.g., religion and positive school experiences for 2017). Commentators describe both resilience and desist- females but not for males) these diferences are not statisti- ance as dynamic and cumulative processes involving a series cally signifcant (Hartman et al. 2009). Nevertheless, there of positive repercussions or chain reactions (e.g., Goldstein is a clear need for robust evaluations of interventions aimed and Brooks 2013). Reductions in severity of ofence may at female youth. In the U.K., of the 11,900 frst-time entrants therefore form the beginning of the desistance process in into Youth Justice Services in England and Wales 2142 were some populations. female (Youth Justice Statistics 2018–2019). In the U.S.A in 2015, of the 884,900 individuals going through juvenile Need for Cultural Tailoring courts, 244,000 were female (Ehrmann et al. 2019).

With regards to race, fndings overall suggest that individ- Counterproductive “One‑Size‑Fits‑All” Nature ual ethnic needs are largely unaddressed. In support of this of Interventions observation, a 10-year review on evidence-based interven- tions for ethnic minorities found that very few studies ana- Across the studies identifed, intervention length and type lyzed the efects of cultural tailoring on program engage- of psychological assessments used varied considerably. In ment, outcomes, and mechanisms of change (Pina et al. other words, most studies are standalone programs with little 2019). In the present review, only 1 out of the 14 studies common ground. It is also clear interventions generally used discussed the importance of cultural sensitivity and provided a “one-size fts-all” approach in that everyone received the training for staf (Hoogsteder et al. 2018). Of the six studies same intervention (more or less). Such approaches may be that included race as a variable in their analysis, only two counter-productive for youth who have potentially diverse reported on those fndings. First, evaluation of the HIT inter- needs, where some may beneft more from individualized vention demonstrated it was more successful with non-white interventions in line with any specifc developmental needs. youth (predicted re-ofending was at 42.7% while actual re- For example, CBT programs (apart from those including ofending was at 16.7%) (Farrington et al. 2002). Second, post-intervention support) are generally more successful assessment of Re-ART, shows that race did not moderate its for adults, perhaps because cognitive abilities are generally success (Hoogsteder et al. 2018). more developed in this population (Mitchell and Palmer Therefore, researchers should consider individual eth- 2004). Youth may beneft more from interventions which nic needs in future interventions. This becomes especially include a focus on skills normally developed in adoles- important when considering the disproportionate numbers cence, for example perspective taking and abstract thinking of black youth in the youth justice system. For example, the (Blakemore and Choudhury 2006). number of black youths in custody in the UK increased by 6% from 2018 to 2019, accounting for 28% of the total youth Importance of Independent Replication custody population (Youth Justice Statistics 2018–2019). In the U.S.A., while 14% of all youth under 18 are black, 42% A further limitation of the interventions in this review is of boys and 35% of girls are in juvenile facilities are black that program developers evaluate most of the interventions. (Sawyer 2019). Specifcally, of the fourteen studies identifed that success- fully reduced re-ofending, time to re-ofend or violent re- ofending, the program developers were heavily involved in nine (64%) of these. Of the four evaluated by independent researchers, two reported less signifcant fndings overall (Cann et al. 2005; Haines et al. 2015). Hence, replication

1 3 Adolescent Research Review and reliability of the fndings remains uncertain. Independ- Funding The author(s) disclosed receipt of the following fnancial sup- ent replication of interventions is vital to assess program port for the research, authorship, and/or publication of this article: The authors acknowledge fnancial support from the Knowledge Economy reliability (cf. Petrosino and Soydan 2005). Further, through Skills Scholarship (KESS II) and Gwynedd/Isle of Anglesey Youth replication and extension, practitioners can modify and tai- Justice Services to complete this study. lor those interventions towards diferent populations with diverse needs. While the performance of an intervention Compliance with Ethical Standards under ideal and controlled circumstances demonstrates its efcacy, “real-world” conditions (i.e., when the study is not Conflict of interest On behalf of all authors, the corresponding author a demonstration by its developers) enable assessment of states that there is no confict of interest. The collaborating company on this project is the Youth Justice Service for Gwynedd and Isle of overall efectiveness (cf. Fritz and Cleland 2003). Anglesey to which none of the authors have any afliation or fnancial connections.

Open Access This article is licensed under a Creative Commons Attri- Conclusion bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long This study sought to address the research gap concerning as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes development of psychological resilience via interventions, were made. The images or other third party material in this article are and its contribution to reduced re-ofending. No review has included in the article’s Creative Commons licence, unless indicated previously attempted to identify contributing psychologi- otherwise in a credit line to the material. If material is not included in cal changes across diferent studies. Despite some limita- the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will tions, fndings from the fourteen studies reviewed showed need to obtain permission directly from the copyright holder. To view a that increasing psychological resources related to reduced copy of this licence, visit http://creativeco​ mmons​ .org/licen​ ses/by/4.0/​ . re-ofending rates, increased time to re-ofend, or reduced severity of such ofences. Overall, increases in positive afect and coping, and decreases in negative afect and aggression positively related to reduced re-ofending. The mechanisms References by which these changes take place appear to include cog- nitive function such as positive self-concept and reduced Agaibi, C. E., & Wilson, J. P. (2005). Trauma, PTSD, and resilience: A Trauma, Violence, & Abuse, 6 negative attitudes. These processes may be in turn be instru- review of the literature. , 195–216. https​://doi.org/10.1177/15248​38005​27743​8. mental in supporting a successful transition into adulthood. Agnew, R. (2001). Building on the foundation of general strain theory: For example, a positive self-identity along with reduced Specifying the types of strain most likely to lead to crime and antagonism may enable the attainment of goals legitimately delinquency. Journal of Research in Crime & Delinquency, 4, through the cultivation of mutually benefcial pro-social 319–362. https​://doi.org/10.1177/00224​27801​03800​4001. Agnew, R., & White, H. R. (1992). An empirical test of gen- community networks. Going forward, tailoring individu- eral strain theory. Criminology, 30, 475–500. https​://doi. alized interventions toward cognitive training, while also org/10.1111/j.1745-9125.1992.tb011​13.x. focusing on the development of inter-personal skills, oppor- Andershed, H., Köhler, D., Eno Louden, J., & Hinrichs, G. (2008). tunities for learning and identifcation of personal strengths Does the three-factor model of psychopathy identify a prob- lematic subgroup of young ofenders? International Journal and support systems seems fruitful. Although individual- of Law and Psychiatry, 31, 189–198. https​://doi.org/10.1016/j. ized approaches may be more costly and more time consum- ijlp.2008.04.003. ing than the one size fts all approaches seen in the current Antonovsky, A. (1991). Halsans mysterium [The mystery of health]. Stockholm: Natur och Kultur. review, the long-term efects may just be transformational. Armour, C. (2012). Mental health in prison: A trauma perspective on importation and deprivation. International Journal of Criminol- Acknowledgements We would like to thank Stephen Wood the Service ogy and Sociological Theory., 5, 886–894. Manager for Youth Justice Service for Gwynedd and Isle of Anglesey Armstrong, S., & Weaver, E. (2013). Persistent punishment: user views (SW) for agreeing to part fund this research project. of short prison sentences. Howard Journal of Criminal Justice, 52, 285–305. Author Contributions SB conceived the idea for the systematic review, Association of Directors of Children’s Services Ltd. (2019). ADCS supervised the study, guided the analysis of the data and contributed Discussion paper: Serious youth violence and knife crime.https​ to the writing and editing of the manuscript; RH designed and con- ://adcs.org.uk/asset​s/docum​entat​ion/ADCS_Discu​ssion​_Paper​ ducted the study, analyzed the data and drafted the manuscript; RR _on_Serio​us_Youth​_Viole​nce_and_Knife​_Crime​_FINAL​.pdf. participated in the design, coordination, and editing of the study; LH *Bahr, S. J., Cherrington, D. J., & Erickson, L. D. (2015). An evalu- participated in the design and coordination of the study. All authors ation of the impact of goal setting and cell phone calls on read and approved the fnal manuscript. juvenile pe-arrests. International Journal of Ofender Therapy and Comparative Criminology, 60, 1816–1835. https​://doi. org/10.1177/03066​24X15​58854​9.

1 3 Adolescent Research Review

Barry, M. (2006). Youth ofending in transition: The search for social relations. Sociology of Health and Illness, 39, 941–958. https​:// recognition. Abingdon: Routledge. doi.org/10.1111/1467-9566.12549​. Blakemore, S. J., & Choudhury, S. (2006). Development of the ado- Diaz, R. (2005). Young people and homelessness. https://engla​ nd.shelt​ ​ lescent brain: implications for executive function and social er.org.uk/__data/asset​s/pdf_fle/0009/48627​/Facts​heet_Young​ cognition. Journal of Child Psychology and Psychiatry, _Peopl​e_and_Homel​essne​ss_Nov_2005.pdf. and Allied Disciplines, 47, 296–312. https​://doi.org/10.111 Eichhorn, S., Brähler, E., Franz, M., Friedrich, M., & Glaesmer, H. 1/j.1469-7610.2006.01611​.x. (2014). Traumatic experiences, alexithymia, and posttraumatic Boeck, T., Flemming, J., & Kemshall, H. (2008). Social capital, resil- symptomatology: A cross-sectional population-based study in ience and desistance: The ability to be a risk navigator. British Germany. European Journal of Psychotraumatology. https://doi.​ Journal of Community Justice, 6, 5–21. org/10.3402/ejpt.v5.23870.​ Borum, R., Bartel, P., & Forth, A. (2002). Manual for the Structured Ehrmann, S., Hyland, N., & Puzzanchera, C. (2019, April). Girls in Assessment of Violence Risk in Youth (SAVRY). Tampa: Univer- the juvenile justice system. U.S. Department of Justice: Juvenile sity of South Florida. Justice Statistics. https​://ojjdp​.ojp.gov/sites​/g/fles​/xycku​h176/ Bouchard, J., & Wong, J. S. (2017). A jury of their peers: A meta- fles​/pubs/25148​6.pdf. analysis of the efects of teen court on criminal recidivism. Eriksson, M. (2016). The sense of coherence in the salutogenic model Journal of Youth and Adolescence, 46(7), 1472–1487. https​:// of health. In M. B. Mittelmark, S. Sagy, M. Eriksson, et al. doi.org/10.1007/s1096​4-017-0667-7. (Eds.), The handbook of salutogenesis. Cham: Springer. https​:// *Boufard, J. A. & Bergseth, K. J. (2008). The impact of re-entry doi.org/10.1007/978-3-319-04600​-6_11. services on juvenile ofenders’ recidivism. Youth Violence and Evans-Chase, M., & Zhou, H. (2014). A systematic review of the Juvenile Justice, 6, 295–318. https​://doi.org/10.1177/00048​ juvenile justice intervention literature: What it can (and can- 65812​46997​3. not) tell us about what works with delinquent youth. Crime & Bowen, K. L., Morgan, J. E., Moore, S. C., & Van Goozen, S. H. Delinquency, 60, 451–470. https://doi.org/10.1177/00111​ 28712​ ​ M. (2013a). Young ofenders’ emotion recognition dysfunction 46693​1. across emotion intensities: Explaining variation using psycho- Fact Sheet on Juvenile Justice. (2008). United Nations Department of pathic traits, conduct disorder and ofense severity. Journal of Economic and Social Afairs: Youth. www.un.org/devel​opmen​t/ Psychopathologic Behavior Assessment, 36, 60–73. desa/youth​/juven​ile-justi​ce-facts​heet.html. Bowen, S., Sustar, H., Wolstenholme, D., & Dearden, A. (2013b). Farrington, D. P. (2000). Explaining and preventing crime: The glo- Engaging teenagers productively in service design. International balization of knowledge—The American Society of Criminology Journal of Child-Computer Interaction, 1, 71–81. https​://doi. 1999 presidential address. Criminology, 38, 1–24. https​://doi. org/10.1016/j.ijcci​.2014.02.001. org/10.1111/j.1745-9125.2000.tb008​81.x. *Burraston, B. O., Bahr, S. J., & Cherrington, D. J. (2014). Reducing *Farrington, D. P., Ditchfeld, J., Hancock, G., Howard, P., Jollife, juvenile delinquency with automated cell phone calls. Interna- D., Livingston, M. S., & Painter, K. A. (2002, April). Evalua- tional Journal of Ofender Therapy and Comparative Criminol- tion of two intensive regimes for young ofenders. Home Ofce. ogy, 58, 522–536. https​://doi.org/10.1177/03066​24X13​48094​7. https​://www.surrey.ac.uk/law/staf​ ​-prof​les/docum​ents/homeo​ Burraston, B. O., Cherrington, D. J., & Bahr, S. J. (2012). Reducing fce​young​ofen​der.pdf. juvenile recidivism with cognitive training and a cell phone fol- Fletcher, D., & Scott, M. (2010). Psychological stress in sports low-up: An evaluation of the RealVictory Program. International coaches: A review of 16 concepts, theory and research. Journal Journal of Ofender Therapy and Comparative Criminology, 56, of Sports Sciences, 28, 127–137. https://doi.org/10.1080/02640​ ​ 61–80. https​://doi.org/10.1177/03066​24X10​38863​5. 41090​34062​08. Butts, J. A., Buck, J., & Coggeshall, M. B. (2002). The impact of teen Ford, J. D., Chapman, J. F., Pearson, G., Borum, R., Hawke, J., & court of young ofenders. Washington, D.C: Urban Institute Wolpaw, J. M. (2008). MAYSI–2 factor structure, reliability, Press. and predictive validity in juvenile detention. Journal of Psy- *Caldwell, M., Skeem, J., Salekin, R., & Van Rybroek, G. (2006). chopathology & Behavioral Assessment, 30, 87–99. Treatment response of adolescent ofenders with psychopathy *Ford, J. D., & Hawke, J. (2012). Trauma afect regulation psy- features: A 2-year follow-up. Criminal Justice and Behavior, 33, choeducation group and milieu intervention outcomes in 571–596. https​://doi.org/10.1177/00938​54806​28817​6. juvenile detention facilities. Journal of Aggression, Maltreat- *Caldwell, M. F., & Van Rybroek, G. J. (2001). Efcacy of a decom- ment & Trauma, 21, 365–384. https​://doi.org/10.1080/10926​ pression treatment model in the clinical management of violent 771.2012.67353​8. juvenile offenders. International Journal of Offender Ther- *Forgays, D. K., & DeMilio, L. (2005). Is Teen Court efective for apy and Comparative Criminology, 45, 469–477. https​://doi. repeat ofenders? A test of the restorative justice approach. Inter- org/10.1177/03066​24X01​45400​6. national Journal of Ofender Therapy & Comparative Criminol- *Cann, J., Falshaw, L., & Friendship, C. (2005). Understand- ogy, 49, 107–118. https​://doi.org/10.1177/03066​24X04​26941​1. ing ‘what works’: Accredited cognitive skills programmes Forth, A. E., Kosson, D., & Hare, R. D. (2003). Psychopathy Checklist- for young offenders. Youth Justice, 5, 165–179. https​://doi. Youth Version. Toronto, ON: Multi-Health Systems. org/10.1177/14732​25405​00500​303. Fritz, J. M., & Cleland, J. (2003). Efectiveness versus efcacy: More Carr, M. B., Vandiver, T. A., & Trish, A. (2001). Risk and protective than a debate over language. The Journal of Arthopaedic and factors among youth ofenders. Adolescence, 36, 409–426. Sports Physical Therapy, 33, 163–165. https​://doi.org/10.2519/ Committee on Child Maltreatment Research, Policy, and Practice jospt​.2003.33.4.163. for the Next Decade: Phase II; Board on Children, Youth, and Gibson, R. A., & Clarbour, J. (2017). Factorial structure of the Resil- Families; Committee on Law and Justice; Institute of Medicine; iency Scale for Children and Adolescents (RSCA) among incar- National Research Council; Petersen, A. C., Joseph, J., & Feit, cerated male adolescent ofenders. Journal of Forensic Practice, M. (Eds.). (2014). New directions in child abuse and neglect 19, 23–36. https​://doi.org/10.1108/JFP-08-2015-0043. research. Washington, DC: National Academies Press. Global Youth Justice. (2018, March 19). Top 25 crimes, ofences and Daykin, N., de Viggiani, N., Moriarty, Y., & Pilkington, P. (2017). violations. (2018). Global Youth Justice. https://www.globa​ lyout​ ​ Music-making for health and wellbeing in youth justice set- hjust​ice.org/resou​rces/top-25-crime​s/. tings: mediated afordances and the impact of context and social

1 3 Adolescent Research Review

Goldstein, S., & Brooks, R. B. (Eds.). (2013). Resilience in children Lipsey, M. W. (2009). The primary factors that characterize efective (2nd ed.). New York: Springer. interventions with juvenile ofenders: A meta-analytic overview. *Haines, A., Lane, S., McGuire, J., Perkins, E., & Whittington, R. Victims & Ofenders, 4, 124–147. https://doi.org/10.1080/15564​ ​ (2015). Ofending outcomes of a mental health youth diversion 88080​26125​73. pilot scheme in England. Criminal Behaviour & Mental Health, Lui, J. H. L., Barry, C. T., & Marcus, D. K. (2019). A short-term inter- 25, 126–140. https​://doi.org/10.1002/cbm.1916.s. vention for adolescents with callous-unemotional traits and emo- Harter, S. (1985). Manual for the self-perception profle for adoles- tion-processing defcits. Journal of Social & Clinical Psychol- cents. Denver: University of Denver Press. ogy, 38, 475–500. https​://doi.org/10.1521/jscp.2019.38.6.475. Hartman, J. L., Turner, M. G., Daigle, L. E., Exum, M. L., & Cul- Luthar, S. S., Cicchetti, D., & Becker, B. (2000). The construct of len, F. T. (2009). Exploring the gender diferences in protective resilience: A critical evaluation and guidelines for future work. factors: Implications for understanding resiliency. International Child development, 71, 543–562. https​://doi.org/10.1111/1467- Journal of Ofender Therapy and Comparative Criminology, 53, 8624.00164​. 249–277. https​://doi.org/10.1177/03066​24X08​32691​0. Mallet, C. A., & Tedor, N. K. (2018). Juvenile delinquency: Pathways Hatcher, R. M., McGuire, J., Bilby, C. A. L., Palmer, E. J., & Hollin, and prevention. Thousand Oaks, CA: Sage. C. R. (2012). Methodological considerations in the evaluation of Marrow, M. T., Knudsen, K. J., Olafson, E., & Bucher, S. E. (2012). ofender interventions: The problem of attrition. International The value of implementing TARGET within a trauma-informed Journal of Ofender Therapy & Comparative Criminology, 56, juvenile justice setting. Journal of Child & Adolescent Trauma, 447–464. https​://doi.org/10.1177/03066​24X11​40327​1. 5, 257–270. https​://doi.org/10.1080/19361​521.2012.69710​5. Höjdahl, T., Magnus, J. H., Mdala, I., Hagen, R., & Langeland, E. McGuire, J. (2008). A review of efective interventions for reducing (2015). Emotional distress and sense of coherence in women aggression and violence. Philosophical Transactions of the Royal completing a motivational program in fve countries. A prospec- Society of London B: Biological Sciences, 363, 2577–2597. https​ tive study. International Journal of Prisoner Health, 11, 169– ://doi.org/10.1098/rstb.2008.0035. 182. https​://doi.org/10.1108/IJPH-10-2014-0037. Ministry of Justice and Youth Justice Board. (2016). Youth custody Hoogsteder, L. M., Hendriks, J., Van Horn, J., & Wissink, I. (2012). report: September 2016.https​://www.gov.uk/gover​nment​/stati​ Responsive aggression regulation therapy: An evaluation study stics​/youth​-custo​dy-data. in a juvenile justice institution. Orthopedagogiek: Onderzoek En Mitchell, J., & Palmer, E. J. (2004). Evaluating the “Reasoning Praktijk, 51, 481–493. and Rehabilitation” program for young ofenders. Journal of *Hoogsteder, L. M., Stams, G. J. J. M., Schippers, E. E., & Bonnes, Ofender Rehabilitation, 39, 31–45. https​://doi.org/10.1300/ D. (2018). Responsive aggression regulation therapy (Re- J076v​39n04​_03. ART): An evaluation study in a Dutch juvenile justice institu- Moftt, T. E. (2003). Life-course-persistent and adolescence-limited tion in terms of recidivism. International Journal of Ofender antisocial behavior: A 10-year research review and a research Therapy & Comparative Criminology, 62, 4403–4424. https://​ agenda. In B. B. Lahey, T. E. Moftt, & A. Caspi (Eds.), Causes doi.org/10.1177/03066​24X18​76126​7. of conduct disorder and juvenile delinquency (pp. 49–75). New *Hubble, K., Bowen, K. L., Moore, S. C., & Van Goozen, S. H. York: The Guilford Press. M. (2015). Improving negative emotion recognition in young Mohajer, N., & Earnest, J. (2009). Youth empowerment for the most ofenders reduces subsequent crime. PLoS ONE, 10, 1–13. vulnerable: A model based on the pedagogy of Freire and experi- https​://doi.org/10.1371/journ​al.pone.01320​35. ences in the feld. Health Education, 109, 424–438. https​://doi. Infurna, F. J., Rivers, C. T., Reich, J., & Zautra, A. J. (2015). Child- org/10.1108/09654​28091​09848​34. hood trauma and personal mastery: Their infuence on emo- Moher, D., Liberati, A., Tetzlaf, J., Altman, D. G., Altman, D., Antes, tional reactivity to everyday events in a community sample G., et al. (2009). Preferred reporting items for systematic reviews of middle-aged adults. PLoS ONE, 10, e0121840. https​://doi. and meta-analyses: The PRISMA statement. PLoS Medicine. org/10.1371/journ​al.pone.01218​40. https​://doi.org/10.1371/journ​al.pmed.10000​97. Johns, D., Williams, K. S., & Haines, K. (2018). A study of prolifc Möller, C., Falkenström, F., Larsson, M. H., & Holmqvist, R. (2014). ofending by young people in Wales, 2009–2015. Wales: Welsh Mentalizing in young ofenders. Psychoanalytic Psychology, 31, Centre for Crime and Social Justice. 84–99. https​://doi.org/10.1037/a0035​555. Jolliffe, D., Farrington, D. P., & Howard, P. (2013). How long MST Services (2018, November 1). Do we know the full extent of juve- did it last? A 10-year reconviction follow-up study of high nile recidivism? [Blog post]. https​://info.mstse​rvice​s.com/blog/ intensity training for young ofenders. Journal of Experimen- juven​ile-recid​ivism​-rates​. tal Criminology, 9, 515–531. https​://doi.org/10.1007/s1129​ Mukherjee, S., & Kumar, U. (2017). Psychological resilience: A con- 2-013-9191-2. ceptual review of theory and research. In U. Kumar (Ed.), The Lansimies, H., Pietila, A., & Husu, S. H. (2017). A systematic review of Routledge international handbook of psychosocial resilience. adolescents’ sense of coherence and health. Scandinavian Jour- New York: Routledge. nal of Caring Sciences, 31, 651–661. https​://doi.org/10.1111/ National Academies of Sciences, Engineering, and Medicine (2019). scs.12402​. The promise of adolescence: Realizing opportunity for all *Lindblom, S., Eriksson, L., & Hiltunen, A. J. (2017). Evaluation of youth. Washington, DC: The National Academies Press. the cognitive intervention programme “A New Direction” target- 10.17226/25388. ing young ofenders in Sweden. Journal of Scandinavian Studies Petrosino, A., & Soydan, H. (2005). The impact of program developers in Criminology & Crime Prevention, 18, 176–190. https​://doi. as evaluators on criminal recidivism: Results from meta-analyses org/10.1080/14043​858.2017.13075​45. of experimental and quasi-experimental research. Journal of Lindblom, S., Eriksson, L., & Hiltunen, A. J. (2018). Criminality, Experimental Criminology, 1, 435–450. thinking patterns and treatment efects—evaluation of the Swed- Pina, A. A., Polo, A. J., & Huey, S. J. (2019). Evidence-based psy- ish cognitive intervention programme ‘new challenges’ target- chosocial interventions for ethnic minority youth: The 10-Year ing adult men with a criminal lifestyle. Journal of Scandinavian Update. Journal of Clinical Child & Adolescent Psychology, 48, Studies in Criminology & Crime Prevention, 19, 204–224. https​ 179–202. https​://doi.org/10.1080/15374​416.2019.15673​50. ://doi.org/10.1080/14043​858.2018.15132​02.

1 3 Adolescent Research Review

Roger, D., & Masters, R. (1997). The development and evaluation of support. Journal of Adventure Education & Outdoor Learning. an emotion control training programme for sex ofenders. Legal https​://doi.org/10.1080/14729​679.2020.17361​09. & Criminological Psychology, 2, 51–64. Turner, T. (2015). Legal writing from the ground up: Process, prin- Roger, D., & Najarian, B. (1989). The construction and validation of ciples, and possibilities. Alphen aan den Rijn: Wolters Kluwer. a new scale for measuring emotion control. Personality & Indi- U. S. Department of Justice. (2015). About violent gangs. https://www.​ vidual Diferences, 8, 845–853. justi​ce.gov/crimi​nal-ocgs/about​-viole​nt-gangs​. Rutter, M. (2006). Implications of resilience concepts for scientifc Vitopoulos, N. A., Peterson-Badali, M., & Skilling, T. A. (2012). The understanding. Annals of the New York Academy of Sciences, relationship between matching service to criminogenic need and 1094, 1–12. https​://doi.org/10.1196/annal​s.1376.002. recidivism in male and female youth: Examining the RNR prin- Salekin, R. T. (2015). Forensic evaluation and treatment of juveniles: ciples in practice. Criminal Justice & Behavior, 39, 1025–1041. Innovation and best practice. Washington, DC: American Psy- https​://doi.org/10.1177/00938​54812​44289​5. chological Association. Walters, G. D. (1995a). The psychological inventory of criminal think- Sawyer, W. (2019). Youth Confnement: The Whole Pie 2019. Prison Policy ing styles, part I. Reliability and preliminary validity. Criminal Initiative. https​://www.priso​npoli​cy.org/repor​ts/youth​2019.html. Justice & Behaviour, 22, 307–325. Serious Crime Act, Chapter 1. (2007). https://www.legis​ latio​ n.gov.uk/​ Walters, G. D. (1995b). The psychological inventory of criminal think- ukpga​/2007/27/conte​nts. ing styles, part II. Identifying simulated response sets. Criminal Shields, I. W., & Simourd, D. (1991). Predicting predatory behavior in Justice & Behaviour, 22, 437–445. a population of incarcerated young ofenders. Criminal Justice Walters, G. D. (1996). The psychological inventory of criminal think- & Behavior, 18, 180–194. ing styles, part III. Predictive validity. International Journal of Sirriyeh, R., Lawton, R., Gardner, P., & Armitage, G. (2012). Review- Ofender Therapy & Comparative Criminology, 40, 105–112. ing studies with diverse designs: The development and evalua- Walters, G. D. (2002). The psychological inventory of criminal tion of a new tool. Journal of Evaluation in Clinical Practice, thinking styles (PICTS). Assessment, 9, 278–291. https​://doi. 18, 746–752. https://doi.org/10.1111/j.1365-2753.2011.01662​ .x​ . org/10.1177/10731​91102​00900​3007. Snow, P. C., Woodward, M., Mathis, M., & Powell, M. B. (2015). Wiebush, R. G., Wagner, D., McNulty, B., Wang, Y., & Le, T. (2005). Language functioning, mental health and alexithymia in incar- Implementation and outcome evaluation of the intensive after- cerated young offenders. International Journal of Speech- care program: Final report. Washington, DC: National Council Language Pathology, 18, 20–31. https​://doi.org/10.3109/17549​ on Crime and Delinquency. 507.2015.10812​91. Wilson, H. A., & Hoge, R. D. (2013). The efect of youth diversion pro- St. James-Roberts, I., Greenlaw, G., Simon, A., & Hurry, J. (2005). grams on recidivism: A meta-analytic review. Criminal Justice & National Evaluation of Youth Justice Board Mentoring Schemes Behavior, 40, 497–518. https://doi.org/10.1177/00938​ 54812​ 45108​ 9​ . 2001 to 2004. London: Youth Justice Board. Wolf, K. T., & Baglivio, M. T. (2017). Adverse childhood experi- Steinberg, L., Chung, H. L., & Little, M. (2004). Re-entry of young ences, negative emotionality, and pathways to juvenile recidi- ofender from the justice system: A developmental perspec- vism. Crime & Delinquency, 63, 1495–1521. https​://doi. tive. Youth violence and juvenile justice, 2, 21. https​://doi. org/10.1177/00111​28715​62746​9. org/10.1177/15412​04003​26004​5. Wood, D., Crapnell, T., Lau, L., Bennett, A., Lotstein, D., Ferris, M., Stoddard, S. A., Whiteside, L., Zimmerman, M. A., Cunningham, R. M., et al. (2018). Emerging adulthood as a critical stage in the life Chermack, S. T., & Walton, M. A. (2013). The relationship between course. In N. Halfon, C. B. Forrest, R. M. Lerner, & E. M. Faust- cumulative risk and promotive factors and violent behavior among man (Eds.), Handbook of life course health development (pp. urban adolescents. American Journal of Community Psychology, 123–143). Cham: Springer. 51, 57–65. https​://doi.org/10.1007/s1046​4-012-9541-7. Youth Justice Board for England and Wales. (2018). Youth Justice *Strom, K. J., Hendrix, J. A., Dawes, D., & Anderson, S. H. (2017). Board for England and Wales—Strategic Plan 2018–21 (pp. An outcome evaluation of the Methodist Home for Children’s 1–13). https://www.gov.uk/gover​ nment​ /publi​ catio​ ns/youth​ -justi​ ​ value-based therapeutic environment model. Journal of Experi- ce-board​-for-engla​nd-and-wales​-strat​egic-plan-2018-21. mental Criminology, 13, 101–124. https://doi.org/10.1007/s1129​ ​ Youth Justice Statistics. (2018–2019). Youth Justice Legal Center, Feb- 2-016-9275-x. ruary 2020. https​://yjlc.uk/youth​-justi​ce-stati​stics​-2018-2019/. Thornton, D. (1987). Assessing custodial adjustment. In B. J. McGurk, Youth justice facts and fgures. (2020). Beyond Youth Custody. https://​ D. M. Thornton, & M. Williams (Eds.), Applying psychology to www.beyon​dyout​hcust​ody.net/about​/facts​-and-stats​/. imprisonment (pp. 445–462). London: HMSO. Trundle, G., & Hutchinson, R. (2020). The phased model of adventure Publisher’s Note Springer Nature remains neutral with regard to therapy: trauma-focussed, low arousal, & positive behavioural jurisdictional claims in published maps and institutional afliations.

Afliations

Rowan Hodgkinson1 · Stuart Beattie1 · Ross Roberts1 · Lew Hardy1

Stuart Beattie 1 Institute for the Psychology of Elite Performance, [email protected] School of Sports, Health and Exercise Science, Bangor University, The George Building, Holyhead Rd., Bangor, Ross Roberts Gwynedd LL57 2PZ, UK [email protected] Lew Hardy [email protected]

1 3