Current trends in the rehabilitation of juvenile offenders

Foreword | The Institute's recent work on adult male offenders has found that the most serious and persistent adult offenders had been detained as a juvenile (see Trends & issues no 267). In terms of crime reduction, interventions that focus on reducing the likelihood of juveniles escalating to adult offenders will have significant benefits for the whole of the Australian community. Research conducted in juvenile justice settings around the world consistently shows that young people who come to the attention of criminal justice agencies have multiple problems and experience high levels of need across all areas of functioning. In meeting these needs, correctional agencies have been increasingly influenced by the model of rehabilitation known as the 'what works' approach. This paper outlines a case management framework for rehabilitating juvenile offenders that includes three of the most important 'what works' principles, namely the risk principle, the needs principle and the responsivity principle. In the longer term, the implementation of the framework will need to be evaluated to determine what works and what doesn't with rehabilitating juveniles.

Toni MakkaiActing Director

Rates of reoffending among juvenile offenders are cause for concern for those involved in criminal justice agencies around the world. In North America, for example, the recidivism rate for young people leaving custody has been reported to be as high as 96 per cent (Lewis et al. 1994). In another study, 88 per cent of British males between 14 and 16 years reoffended within two years of release from custody (Hagell 2002). Reoffending among juveniles following community orders appears to be much lower, but the majority still reoffend. In Australia, a Victorian government study into recidivism among juvenile justice clients (DHS 2001) reported that nearly half (41%) of a sample of more than 1,500 juvenile justice clients reoffended, with this rate rising to 61 per cent for those who had previously been on supervised orders.

Such statistics provide a strong rationale for juvenile justice services to scrutinise their models of service delivery and maximise the effectiveness of their rehabilitation programs. It is encouraging that effective rehabilitation programs are available. (The term 'program' is used in this paper to refer to psychological treatments designed to reduce recidivism. This paper does not discuss other types of programs, such as those administered by the police and courts - for example, diversionary programs.)

In their review of more than 200 programs delivered to serious and violent young offenders, Lipsey and Wilson (1998: 338) reported that the best programs were capable of reducing recidivism rates by as much as 40 per cent. They regarded this as an 'accomplishment of considerable practical value in terms of the expense and social damage associated with the delinquent behaviour of these juveniles.' A reduction in recidivism of this magnitude compares favourably with those commonly cited in reviews of rehabilitation programs for adult offenders which have found that these programs typically reduce rates of reoffending by between 5 and 18 per cent (Hollin 1999). Effectiveness rates are known to be higher in the best quality programs (Andrews & Bonta 1998). Typically these programs have a strong theoretical basis, employ a structured behavioural or cognitive-behavioural approach that focuses on the attitudes and beliefs of offenders, and are delivered by well trained staff.

What works

In recent years, correctional agencies around Australia have been increasingly influenced by the model of rehabilitation commonly known as the 'what works' approach to offender rehabilitation. This approach has been described in more detail elsewhere (Howells & Day 1999; Day & Howells 2002) but can be summarised by a core set of principles of human service delivery. Collectively, these principles suggest that reductions in recidivism can be maximised when programs select appropriate candidates, target factors that directly relate to their offending, and are delivered in ways that facilitate learning (Table 1). Programs which adhere to these principles target those at the highest risk of reoffending. They offer high-risk offenders the most intensive interventions and seek to change factors that are known to be directly related to the reasons for offending, such as antisocial attitudes, substance use and anger.

In recent years juvenile justice agencies around Australia have become increasingly interested in applying these core principles to programs delivered with juvenile offenders. If adhering to these principles significantly improves the effectiveness of programs for adult offenders in reducing recidivism, then they may also improve the outcomes of juvenile justice programs. This paper discusses some of the issues that arise in applying three of the most important 'what works' principles to juvenile justice clients.

Risk principle

The first, and arguably the most important, principle in the 'what works' model is the 'risk principle'. This suggests that offenders most likely to reoffend should receive the most intensive rehabilitation. In practice, adhering to the risk principle involves using a risk assessment tool to classify each offender into a low, medium or high risk group. More intensive programs are offered to those assessed as medium or high risk. An issue that arises in applying the risk principle to juvenile offenders in custody is the high base-rate with which young people offend (see earlier). If an overwhelming majority of juveniles released from juvenile detention centres commit any other offence, then a risk assessment is unlikely to discriminate between different groups. Rather, applying the risk principle in juvenile justice settings aims to identify those who will go on to offend seriously or persistently and then offer them intensive rehabilitation.

Table 1: 'What works' principles for effective offender rehabilitation
The risk principle Published research identifies variables associated with the likelihood of an individual reoffending. These risk principles include those not amenable to intervention (static risk factors), and those that might change over time (dynamic risk factors). Static risk factors include age of onset of crime, offence history and family structure. Research suggests that higher risk offenders will benefit the most from rehabilitation interventions and that the intensiveness of services delivered should be proportional to the level of risk.
The needs principle The term 'criminogenic needs' refers to risk factors that are dynamic or amenable to change through intervention. The needs principle suggests interventions should target needs of this sort, as they are most directly related to recidivism. Examples of criminogenic needs that form important targets for intervention with young offenders are drug and alcohol use, anger and violence problems, and beliefs or attitudes that support offending.
The responsivity principle The responsivity principle focuses on client and program characteristics that influence the offender's ability to learn in a therapeutic situation. Treatment is a learning experience and individual factors that interfere with, or facilitate, learning are termed responsivity factors. These factors can also be understood as contextual variables, which may influence treatment outcome. These variables make a difference to the skills, strategies or identities that individuals develop and to the support available when transitions are made. Factors such as age, ethnicity, gender, disability and socioeconomic status can be considered key responsivity factors.
The integrity principle In contrast to the demands made by the responsivity principle to individualise interventions, an important component of quality assurance is to emphasise the need for program integrity. Program integrity refers to the extent to which an intervention program is delivered in practice as intended in theory and design.
The professional discretion principle The principle of professional discretion allows for professionals to make decisions on characteristics and situations not covered by the preceding principles. It makes sense to build scope for professional judgment into any rehabilitation system, rather than rely upon rigid administration of static principles.
Source: Summarised from Andrews & Bonta (1998)

Longitudinal research increasingly identifies a small subgroup of juvenile offenders who begin their offending careers early, commit more offences (and more serious and violent offences) and account for a disproportionate number of offences in their adult years (Loeber & Farrington 1998). Members of this subgroup are sometimes referred to as life-course persistent offenders, and are contrasted with those for whom offending might be considered as adolescence-limited (it diminishes as maturation occurs). Moffitt (1994: 12) described life-course persistent offenders as those who:

...exhibit changing manifestations of antisocial behaviour: biting and hitting at four, shoplifting and truancy at ten, selling drugs and stealing cars at 16, robbery and rape at 22, and fraud and child abuse at 30. The underlying disposition remains the same, but its expression changes form as new social opportunities arise at different points in development.

It has been estimated that while approximately only five to six per cent of juvenile offenders would be included in this category, this group would be responsible for the majority of crimes (Moffitt 1993). In applying the risk principle to juvenile clients, we suggest that the most intensive programs should be offered to those most likely to engage in serious and/or persistent reoffending. Although some researchers are seeking to find ways to reliably identify this group, our knowledge base does not appear to allow for the accurate and reliable prediction of which young people will go on to offend seriously and persistently. Given the obvious dangers associated with self-fulfilling prophecies, labelling young people as high risk at an early age needs to handled carefully and responsibly.

Needs principle

The second core principle of the 'what works' approach - the needs principle - suggests that the most effective programs are ones that intervene to change needs most directly related to offending. These are commonly known as criminogenic needs. Meta-analysis was used by Cottle, Lee and Heilbrun (2001) to combine the results of 24 studies conducted with juvenile offenders involving more than 15,000 participants. They found that five main categories of risk factors predicted juvenile reoffending, four of which could be considered as dynamic risk factors or areas of criminogenic need. These were:

  • family and social factors (for example, significant family problems; ineffective use of leisure time; delinquent peers);
  • educational factors;
  • substance use history; and
  • non-severe mental health problems.

Programs that adhere to the needs principle focus on changing these factors. While it may be acceptable to only intervene with these offence-related needs in adults, juvenile justice is concerned with the much broader responsibility to care for and nurture children, to ensure a successful transition into adult life. Juvenile justice services are not only concerned with addressing needs directly related to the risk of reoffending, they are also concerned with addressing a broad range of needs that centre around risks, in addition to the wellbeing of children and their families, child abuse and neglect, truancy, substance abuse, and mental health problems.

Research conducted in juvenile justice settings around the world consistently shows that young people have multiple problems and experience high levels of need across all areas of functioning. For example, a young person who has an offending background, family problems and substance use, and who is disengaged from school, needs support in all areas, not just in desisting from offending. Narrow offending-targeted responses may, by themselves, be insufficient to meet this range of needs. The relationship between criminogenic programs and programs that seek to meet non-criminogenic needs has been the subject of some debate in adult rehabilitation literature (for example, Ward & Brown 2004) and is critical to the development of programs for juveniles.

A vital task in effective rehabilitation programming is to ensure the distinctive needs of client groups are determined and addressed. It was argued by Alder (1997: 2) that despite their significance as a group, 'girls are still barely visible in our theories, research and policy documents in juvenile justice.' It is difficult to access data and research that identifies gender-specific risk markers or makes clear statements about criminogenic need. However, the vulnerabilities of young women in juvenile justice lead many to look towards integrated models of intervention that address multiple problems and high levels of need.

Responsivity principle

The third major principle of the 'what works' approach is the 'responsivity principle'. This principle suggests that the most effective programs match the learning styles of offenders, such that they are actively engaged in a process of behavioural change. Engaging young people in programs is likely to be difficult, given that adolescents are generally unlikely to seek help from professional services and hold negative attitudes towards criminal justice agencies. A series of focus groups with young offenders in the UK (Lyon, Dennison & Wilson 2000) concluded that although most participants had considerable contact with professionals and organisations in the criminal justice system, these contacts generally proved unsatisfactory (particularly with the police).

Figure 1: Proposed framework for rehabilitation in juvenile justice Proposed framework for rehabilitation in juvenile justice

It seemed unlikely that this group of young offenders would approach criminal justice services for help with personal problems, although they did attach great importance to being treated with a degree of respect appropriate to their age and level of development.

A responsive program would engage young people in activities and learning that are personally meaningful, and would be delivered in a way that makes sense to participants. This may mean, for example, engaging young people in a range of practical or physically based activities and relying less on formal educational methods that require high levels of literacy or concentration. Preferred learning styles will likely change over the course of adolescence, with adult learning becoming increasingly relevant as children move through adolescence. Thus, it is not only likely that risk factors for reoffending and criminogenic needs will change over the course of adolescence, but also that methods of program delivery most likely to engage young people will also change.

A developmental approach

Evaluations of programs based on the 'what works' principles have generally been reported for the older juvenile age group (typically 17 to 21 years old). There have been relatively few evaluations of programs designed for the younger age group (10 to 14 years). Some of the most effective programs are offered to juvenile offenders (Redondo, Garrido & Sanchez-Meca 1998), although this may be because offenders of this age are more likely to receive programs that use the most effective methods (behavioural and cognitive-behavioural methods), or because offending typically diminishes after this age. The Youth Lifestyles Survey in the UK (Nacro 2001) reported that the peak age of offending, at least according to self-report data, is around 18 years for males and 14 years for females, with the highest levels of offending in the 18 to 21 age group. In other words, it may be that programs for this age group are effective because for many young people, offending is adolescent-limited and diminishes as they mature.

The 'what works' approach requires greater adaptation for younger juvenile justice clients (aged 10 to 17). While there has been relatively little discussion in the literature about the impact of developmental processes upon rehabilitation programming, this is a critical issue. Risk factors, criminogenic needs and responsivity issues are all likely to change over the course of adolescence. Adolescence is a period of great change - biologically, psychologically and socially. Biological development, for example, encompasses profound physical changes caused by the onset of puberty. While some evidence suggests that physical changes (such as in hormone levels and the functioning of the endocrine system) are associated with behavioural problems (such as violence and aggression), the amount of variance explained by these changes is thought to be small when compared to the impact of social influences (Weiss & Hawley 2002). Furthermore, it makes little sense to treat biological development separately from the psychological and social development dimensions of adolescence. For example, while there is some evidence linking the early onset of puberty to risk factors, including delinquency for boys and girls, and early maturation is considered a risk factor for offending in young women, this is only apparent when it leads to young women associating with older peers. Psychological development in this context refers to changes in identity and independence from others, as well as beliefs about risk-taking and what is appropriate or acceptable behaviour. Social development would impact upon choice of peer group and the extent to which antisocial behaviour or offending is sub-culturally valued.

Developmental issues are also likely to influence the types of problems juvenile offenders experience. Younger adolescents facing the transition to high school and the onset of puberty, for example, have different developmental needs to older adolescents facing the transition from school to work. Relationships with, and dependency on, family and care-givers are also likely to change with age. Disentangling the effects of maturation in terms of identifying criminogenic needs is not straightforward. Work on violence prevention suggests that:

  • young children (under 8) might benefit most from programs emphasising emotional regulation and parent-child interaction;
  • children in middle childhood (8 to 11) should attend programs focusing on social competence; and
  • early (12 to 14) and middle (15 to 18) adolescents should attend programs that address the development of pro-social peer groups, conflict resolution and work/job skills (Farrell et al. 2001).

An issue of special significance for this developmental period is the emergence of mental health problems and mental disorders.

Case management

In Victoria, juvenile justice services are considering implementing four levels of programming for clients (see Figure 1). The first level of service delivery would focus solely on sentence or order administration and would be offered to low-risk offenders, including those who only require supervision and monitoring or who have family support sufficient to meet their needs. These offenders may also have needs relating to integrating with the community. The needs are generally not considered criminogenic, but are an important part of the work in juvenile justice. Interventions of this sort are at the level 2 stage and typically address issues that might obstruct community integration such as employment, accommodation, education and leisure.

For young offenders at medium to high risk of offending, level 3 or 4 interventions are indicated. The aim of these interventions is to reduce the risk of offending and programs should explicitly target criminogenic needs. A range of programs for different criminogenic needs should be available, including for common areas of need such as substance use, pro-offending attitudes, peer/criminal associate influences and family influence. As a minimum, these needs should be targeted through systematic intervention.

Level 4 interventions are the most intensive and should be offered to the highest risk or most persistent offenders. According to professional judgment, they may also be offered to those whose offences are considered particularly harmful. It is likely that the number of young people requiring level 4 programs will be small, and as such resources might be targeted towards particular offending groups (for example, serious and/or persistent violent and sexual offenders).


This paper has discussed the recent trend for juvenile justice agencies in Australia to review how they approach rehabilitating juvenile offenders in light of the 'what works' model of rehabilitation, which has been highly influential in adult criminal justice agencies. The 'what works' principles suggest a model of differentiated case management, whereby those with the highest level of need and greatest risk of reoffending are given the most intensive programs. This is not only likely to maximise program effectiveness, but also provide a reasonable and empirically defensible way of allocating resources. Although there are differences between adult and juvenile services, we would argue that adherence to each principle is still likely to improve rehabilitation outcomes. A number of issues are identified in applying the three main 'what works' principles to juvenile offenders, the most important of which is developmental changes taking place between the age of entry into juvenile justice services (typically age 10 in Australia) and the age when adult services are offered.

Placing different types of programs within a framework such as that being developed by juvenile justice services in Victoria has the advantage of distinguishing programs that explicitly seek to reduce reoffending from those that seek to increase the wellbeing of young people. It is likely that many current programs commonly delivered to young people in juvenile justice settings across Australia would be considered second and third level under this framework. In many instances, it may be straightforward to adapt or modify such programs to integrate them with the 'what works' model. This may be successful in further developing programs for juveniles who commit serious and/or persistent offences, thereby further reducing rates of juvenile reoffending.


This paper arose out of a review of rehabilitation services conducted for the Juvenile Justice Branch of the Department of Human Services, Victoria, by the Forensic Psychology Research Group, University of South Australia, and the Australian Institute of Criminology. Thanks are extended to two anonymous referees for their helpful comments on a previous draft of this paper.


  • Alder C 1997. Young women and juvenile justice: objectives, frameworks and strategies. Paper presented at the AIC conference Juvenile crime and juvenile justice: toward 2000 and beyond. Adelaide: 26-27 June
  • Andrews DA & Bonta J 1998. The psychology of criminal conduct. Cincinnati: Anderson
  • Cottle CC, Lee RJ & Heilbrun K 2001. The prediction of criminal recidivism in juveniles: a meta-analysis. Criminal justice and behaviour vol 28: 367-94
  • Day A & Howells K 2002. Psychological treatments for rehabilitating offenders: evidence-based practice comes of age. Australian psychologist vol 37: 39-47
  • Department of Human Services (DHS) 2001. Recidivism among Victorian juvenile justice clients 1997-2001. Victoria: Department of Human Services
  • Farrell AD, Meyer AL, Kung EM & Sullivan TN 2001. Development and evaluation of school-based violence prevention programs. Journal of clinical child psychology vol 30: 207-20
  • Hagell A 2002. The mental health of young offenders - bright futures: working with vulnerable young people. London: Mental Health Foundation
  • Hollin CR 1999. Treatment programs for offenders. International journal of law and psychiatry vol 22: 361-72
  • Howells K & Day A 1999. The rehabilitation of offenders: international perspectives applied to Australian correctional systems. Trends & issues in crime and criminal justice no 112. Canberra: Australian Institute of Criminology
  • Lewis DO, Yeager CA, Lovely R, Stein A & Cobham-Porterreal CS 1994. A clinical follow-up of delinquent males: ignored variables, unmet needs and the perpetuation of violence. Journal of the American academy of child and adolescent psychiatry vol 33: 518-28
  • Lipsey MW & Wilson DB 1998. Effective interventions for serious juvenile offenders: a synthesis of research. In R Loeber & D Farrington (eds) Serious and violent juvenile offenders. New York: Sage
  • Loeber R & Farrington D 1998. Serious and violent juvenile offenders. New York: Sage
  • Lyon J, Dennison C & Wilson A 2000. Messages from young people in custody: focus group research. Research findings no 127. London: Home Office
  • Moffit TE 1993. Adolescent limited and life course persistent antisocial behavior: a developmental taxonomy. Psychological review vol 100: 674-701
  • Moffit TE 1994. Natural history of delinquency. In EGM Weitekamp & H Kerner (eds) Cross-national longitudinal research on human development and criminal behavior. Boston: Kluwer
  • Nacro 2001. Young adult offenders: a period of transition. London: Nacro
  • Redondo S, Garrido V & Sanchez-Meca J 1998. What works in correctional rehabilitation in Europe: a meta-analytic review. In S Redondo, V Garrido, J Perez & R Barbaret (eds). Advances in psychology and law: international contributions. Berlin: De Gruyter
  • Ward T & Brown M 2004. The good lives model and conceptual issues in offender rehabilitation. Psychology, crime and law vol 10: 243-58
  • Weiss JR & Hawley KM 2002. Developmental factors in the treatment of adolescents. Journal of consulting and clinical psychology vol 70: 21-43

About the authors

Dr Andrew Day and Professor Kevin Howells are members of the Forensic Psychology Research Group at the Centre for Applied Psychological Research, University of South Australia.Debra Rickwood works at the University of Canberra.