Abstract
The focus of this review is the potential role of public sector child and youth mental health services (CYMHS) in the utilization of selective evidence-based programs for the prevention and treatment of aggressive behaviour in young people. An all-inclusive outline of prevention and treatment strategies is avoided as such an endeavour would traverse domains as broad as socio-economic policy and the function of the juvenile justice system, realms well beyond the province of mental health services. Child and youth mental health services are defined as community, multidisciplinary mental health teams often working in close partnership with mental health inpatient units. The core business of such teams is the assessment and treatment of child and youth mental health disorders and problems. As well, CYMHS often function in close cooperation with a variety of child and youth welfare, education and health services. Any proposed augmentation to CYMHS roles needs to take into account both societal concerns with youth crime and the increasing emphasis on evidence-based approaches embedded within recent public sector mental health services reforms.
Societal and service provision context
There are two broad reasons for the involvement of CYMHS in the prevention and treatment of aggressive/ antisocial behaviour. First, CYMHS clients manifesting attention deficit hyperactivity disorder (ADHD), conduct disorder (CD) and oppositional defiant disorder (ODD) as described in DSM–IV [1], are at risk of adolescent and adult antisocial behaviour [2]. A broad research consensus has concluded that the subgroup of juveniles who persistently offend have characteristics of mental health disorders with the onset of behaviour disturbances in the preschool and primary school years [3]. Therefore intervention by CYMHS with the above groups could moderate the development of future antisocial behaviour. Second, aggressive and antisocial behaviour are a major concern and cost to society. Child and youth mental health services can make a positive contribution to the reduction of societal disquiet with this serious problem.
Australia ranks within the lead group of industrialized nations (England, Wales, Netherlands and Sweden) in terms of the percentage of people victimized in the previous year [4]. Australian surveys found that about 50% of people were apprehensive of crime [5]. A Western Australian survey on neighbourhood perceptions of antisocial behaviour found that problems most frequently reported included burglaries, noisy reckless driving and vandalism; all behaviours typical of young offenders [6]. A related school survey described students' concerns about violence [7]. One-third of pupils reported involvement in at least one physical fight during the previous 6 months. Seventeen percent of this group reported that they had four or more fights in the previous 6 months. The combined evidence suggests a high level of community concern and contact with crime and adolescent violence.
An accurate perspective on both the extent of juvenile participation in crime as well as trends over time is difficult, due to variations in legal classifications, policing practice, court processing and official data collection systems. An analysis of 1993/94 Victorian police data concluded that 21% of offences were committed by juveniles [8]. As youths aged 10–18 years were approximately 14% of the population, the above estimate points to a disproportionately high involvement of youth in crime [9]. Within the limits of the data collection systems, juvenile arrest trends from 1973 to 1995, have demonstrated rises in serious assault and robbery while burglary and motor vehicle theft have remained roughly constant over the same time span [8].
There are no contemporary estimates of the cost of youth crime in Australia. A 1990 study of crime committed by youths aged 7–18 years arrived at the figure of $1.5 billion per year [10]. Using economic data available through Economajic.com Time Series, it is estimated that the former amount is now equal to $2 billion in 2001 terms or 2% of the gross national product (GDP) [11]. Much of this cost is due to crimes committed by repeat offenders as exemplified by a Queensland study of the court records of a cohort of young offenders. Almost 66% appeared in court only once and were responsible for only 35% of the court appearances. Repeat offenders accounted for the remaining appearances with 1% of the offenders appearing at least 10 times [12].
In summary it seems reasonable to conclude that Australia has been experiencing a rise in violent juvenile crime as well as a flattening of trends in property crimes. Of concern is the small population base of young people disproportionately contributing to the overall level of detected crime. In addition, juvenile crime imposes a huge cost on the community.
The principles guiding service provision within CYMHS have emerged over the last decade in the form of a National Mental Health Strategy [13–15]. The principles are: mainstreaming of services to promote continuity of care across service components; locally available care through equitable distribution of mental health services; prioritizing services to those most in need; establishing links with the primary health sector; and improving intersectoral links. As well, there is an emphasis on specific clinical and service delivery standards such as the utilization of the most effective treatments. The Second National Mental Health Plan has broadened the focus for mental health service delivery to include promotion/prevention programs [16]. In addition there has been the development of the Mental Health Promotion and Prevention Action Plan [17], which has identified priority target groups across the life span. The report distinguishes between promotion activities that focus on the improvement of wellbeing and prevention activities that aim to reduce the incidence of mental health problems/disorders by either universal, selective or indicated programs (discussed in more detail in next section). The domain of evidence-based mental health promotion in children has been further enhanced by a report by Raphael [18] in which key developmental transitional points are identified as well as a number of strategies that could be utilized. This methodology has much in common with the approach outlined later.
Prevention/treatment programs: principles and problems
The Australian Early Intervention Network for Mental Health in Young People [19] has outlined some of the general definitions and principles applicable to the selection of prevention and treatment programs. The term ‘prevention’ applies to interventions that occur before the onset of a disorder. Universal prevention interventions are provided to the whole population in the absence of identified individual risks. Selective prevention targets individuals who are at high risk of developing mental problems. Indicated prevention programs focuses on high-risk individuals who may have discernible signs of mental health disorders. Interventions with identified cases occur in standard mental health treatment services.
Underscoring the prevention philosophy is the concept of risk reduction or the ‘risk factor prevention paradigm’ as outlined by Farrington [20]. The strategy seeks to identify factors that increase the probability of later antisocial behaviour and intervene to eliminate these risk factors. More than three decades of research has established a consensus on the key risk factors for persistent antisocial behaviour [21]. Some of these include individual child factors such as: obstetric delivery complications; hyperactivity; impulsivity and aggressiveness; learning difficulties; and the family factors such as poor attachment, child maltreatment, poor family management practices and poor parental supervision. The selection of any prevention/treatment program by CYMHS would be premised on the principle of intervening with specific risk factors.
The principles or criteria of evidence-based practice have been outlined by the Centre for the Study and Prevention of Violence (CSPV) [22]. They include strong research design, especially randomised allocation of subjects, evidence of significant deterrence effects, multiple site replication, sustained effects, change in targeted risk behaviours and the availability of cost data. A recent update by the CSPV has identified 11 programs that either reduce adolescent violence and delinquency or predelinquent childhood aggression and conduct disorder [23]. There have been a number of other reviews of this topic by expert researchers and government departments, which have not included as stringent criteria as those utilized by the CSPV [24–26]. This has resulted in a broader menu of programs characterized by high levels of evidence-based criteria. In Australia, the report by the Commonwealth Government's Department of the Attorney General (1999) [27] on crime prevention has outlined key life-cycle transition points such as infancy, early toddler-hood, and primary school and secondary school periods where specific evidence-based prevention/treatment programs may alter the genesis of a delinquent life style. Before outlining some of these programs in detail it is important to consider the problems and limitations of evidence-based mental health practice.
Pincus and Zarin [28] have identified several issues constraining the applicability of evidence-based practice. Some of these problems include insufficiency of available evidence, limitations in relevance of such evidence for clinical and policy development and lack of clarity about how to integrate the evidence into everyday clinical practice. King [29] has stressed the need for caution with evidence-based approaches especially where a ‘narrow hegemony’ can emerge that threatens the interests of both consumers and providers of services. The adoption of evidence-based programs is likely to be further complicated by the presence of a diverse range of intervention approaches for child and adolescent aggression within CYMHS. These approaches are often unique to the culture of particular CYMHS. Examples of established treatment customs may range from individual, group and family therapies to psycho-pharmacological approaches. It is clear that any attempt by mental health services to introduce new programs needs to be sensitive to the limitations of evidence-based mental health and the current intervention practices established within services.
Evidence-based programs
Three assumptions provide the rationale for the selective choices of programs detailed later. First, only those programs that have been repeatedly endorsed by numerous reviews (especially those selected by CSPV) are discussed. This inevitably limits the selection process. Second, the choices are based on the life-cycle transition points of infancy, early toddler-hood, and primary school and secondary school periods where prevention/ intervention can alter the birth or persistence of an aggressive/antisocial way of life [18], [27]. Third, the choices of evidence-based programs are premised on the author's contemporary understanding of how CYMHS function. It is assumed CYMHS mainly operate during daylight hours and most clients are interviewed in a clinic. Limited home-visiting services are available. The age range of patients is assumed to be from infancy/ toddler-hood to late adolescence. Liaison with other services is a common characteristic of most intervention processes. The programs selected below have characteristics that allow for wholesale adoption within the constraints outlined earlier, or require new partnerships to be developed or have such compelling evidence-based features that significant reform of CYMHS is essential.
Infancy/preschool and early primary school years
Infancy period
The work of David Olds is one of the best examples of high-quality prevention research during the infancy period [22], [30]. The principles of this intervention can be conceptualized as a selective prevention program targeting the early risk factors such as maternal health problems during pregnancy and child abuse/neglect. Despite this study having commenced 25 years ago with the risk that it may be considered outdated, the advantage of its long-term follow-up has provided a model for contemporary interventions. The University of Rochester Nurse Home Visitation Program [30] recruited 400 pregnant high-risk mothers (i.e. either adolescent or unmarried or low-income) who were randomly assigned to one of four levels of intervention. Level one involved information and support on child development and health. Level two included free transportation to prenatal care and wellchild visits added to the level one services. Level three consisted of nurse home visiting, and level four continued home visiting until the child's second birthday. Postnatal visits focused on education concerning infant development, recruitment of family support for the mother and linkage with other services. Results from the multilayered intervention demonstrated an amelioration of adolescent antisocial behaviour, especially in those exposed to level four interventions. The level four groups compared to level one and two interventions had 50% fewer incidences of arrests; a third less incidence of convictions and probation violations and 10% fewer juvenile supervision orders. At 2 years and 15 years, there were changes in the key risk factors for antisocial behaviour such as lower incidence of child abuse and neglect, fewer child behaviour problems due to substance abuse, fewer pregnancies, more mothers returning to work and less criminal behaviour on part of low income unmarried mothers.
The pre-school period
One of the best examples of preschool prevention programs with long-term outcomes measures is the High/Scope Perry Preschool Project [27], [31]. This study is best characterized as a selective prevention program targeting the risk factors of low educational attainment and weak parental involvement. In this study, children 3–4 years, with low intellectual performance from low income families were randomly allocated to intervention and non-intervention conditions. The intervention condition involved high quality daily educational classes for the children as well as a home visiting program to encourage parental involvement in the child's education. Parents attended monthly support and information exchange groups. Follow-up of the children and controls between ages 19 and 24 revealed that the preschool intervention group demonstrated less chronic offending (7% vs. 17% for controls), fewer property and violent offences and school dropouts were 20% lower in the intervention group. As with the Olds study [30], the age of this intervention may threaten its relevance to current clinical practice. However, its importance lies in the evidence of cognitive-behavioural enrichment in the preschool years and can have an impact on later delinquent outcomes.
The CSPV has endorsed ‘The Incredible Years Parent Teacher and Child Training Series’ [23], [32], a selective prevention program targeting children between ages 2–8 who are at risk for developing conduct problems. Longterm adolescent/adult outcomes are not available. The risk factors targeted include child problem-solving and poor parenting. The elements of the program consist of three levels of parent training (BASIC, ADVANCE and SCHOOL), a teacher training program and a childtraining module. BASIC provides the central aspect of the program, which encompasses group teaching of parenting skills such as interactive play, reinforcement, non-violent discipline, logical and natural consequences. The ADVANCE component assists parents with other risk factors including poor anger management, marital discord and depression. The SCHOOL ingredient of the program teaches parents how to consolidate the child's academic and social skills. Teachers are trained in classroom management skills, helping to motivate, praise pro-social behaviour and cooperation in the children. The child-training module includes improving the child's capacity to empathise with others and manage their own anger. The three elements of the program (parent, teacher and child-training) have all demonstrated significant improvements in child behaviour, parenting interactions, classroom management skills and improved child social skills and conflict management approaches. The combination of all three elements has been shown to decrease aggressive behaviour across all social settings (parents, teachers and peers).
The primary school years
The Conduct Problems Prevention Research Group study or ‘Fast Track’ has been nominated as an exceptional prevention program [22], [33]. This program involves long-term multicomponent, multisite interventions applying a randomised control group design to high-risk children in Grade 1. Both universal and selective interventions took place. Risk factors targeted include a range of child factors such as aggression and poor problem solving as well as parenting deficits. In the universal approach all children in a class were exposed to the ‘Promoting Alternative Thinking Strategies’ (PATHS) curriculum throughout the year. PATHS promoted emotional understanding and communication skills, friendship skills, and self-control and social problem-solving skills training. Selective interventions for high-risk groups included parent groups to improve child behaviour, social skills groups for children, and remedial tutoring. The interventions were continued over 3 years, based on the level of functioning of the child and family. Evaluation at the 3-year stage concluded that of those children who meet the criteria for caseness, 37% were problem-free compared to 27% in the control group [34].
Several other programs (best conceptualized as indicative interventions) have demonstrated positive and continuing change in child behaviour, in particular problem-solving and parenting skills interventions during the preschool and primary school period [35], [36]. Wasserman and Miller (1998) [24] have pointed out that successful interventions, such as ‘Fast Track’ and ‘The Incredible Years’, have in common the utilization of parent management training (PMT) and child social competence training (CSCT). Parent management training teaches parents to use more effective discipline, monitor and supervise their children and to reward prosocial behaviour. Child social competence training is based on the evidence that aggressive children have poor capacity to interpret social cues, tend to assume harmful intent in other children's behaviour, believe that aggressive responses are acceptable and have poor social problem-solving skills. child social competence training teaches the child how to solve problems, control anger and interact more effectively with others.
The adolescent years
The programs discussed in this section can best be classified as examples of case identification of established disorders. Two treatment approaches for adolescent antisocial behaviour have been frequently highlighted [22–27].
Functional family therapy (FFT)
Functional family therapy (FFT) is based on the concept that symptoms or behaviour serve a function or have meaning within a family system [37]. Functional family therapy encourages the family to understand the reinforcement systems that are operating to maintain the problem behaviour. Social learning theory principles are utilized as well as examination of family members' cognitive attributions in uncovering the processes of reinforcement. When the family can see alterative ways of understanding the problem behaviour new patterns of reinforcement are tested such as better communication, social problem-solving, and exchange of privileges. Functional family therapy has demonstrated reduction of delinquent behaviour, findings that have been replicated over 25 years. A limitation with FFT has been its use specifically with delinquent youth but not with clinical samples of conduct-disordered youth [22].
Multisystemic therapy (MST)
Multisystemic therapy (MST) has been repeatedly endorsed as one of the most efficacious interventions with aggressive and antisocial adolescents [22–27]. The efficacy of MST has been extensively documented over the last decade [38]. Multisystemic therapy has been tested with a variety of groups, for example inner city delinquents, child-abusing parents and drug-abusing offenders. To date 800 families have been part of MST trials. In comparison with control groups MST has demonstrated outcomes such as decreased long-term re-arrest rates ranging from 25% to 70%. As well, MST effects have been sustained over 2 years with 39% of treated youth not being re-arrested compared to 20% of youth who had usual services. In addition, MST has demonstrated other relevant outcomes compared to control groups, such as improved family relations, school attendance and decreased adolescent psychiatric symptoms. The Washington State Institute for Public Policy [39] published a report on the cost effectiveness of various crime reduction programs. It estimated that the net gain for MST per case was in the range of US$31 661–131 918. Multisystemic therapy was one of the most cost-effective programs reviewed.
The broad aim of MST is to evaluate the multiple socio-ecological settings that the adolescent experiences and to identify the reinforcers of antisocial behaviour that permeate those settings. In addition, the adolescent's strengths are differentiated. Therapy involves a rigorous program of reversing the effects of the reinforcers of antisocial behaviour and promoting the adolescent's strengths across all systems. Multisystemic therapy does not derive its philosophical foundations from a particular theoretical school. Rather, it draws on multiple influences and techniques to understand and intervene in the broad social systems that influence the adolescent. Hypotheses are formulated from the assessment data using only measurable variables. Based on these formulations, interventions are planned, tested and revised in accordance with outcomes. Table 1 outlines most of the characteristics of MST that contribute to its efficacy.
Characteristics of multisystemic therapy
An essential element of MST is its model of clinical supervision. A study comparing the effectiveness of MST versus usual juvenile justice services for violent and chronic juvenile offenders found that, although MST did result in a better outcome than traditional processes, the results were not as efficacious as in previous studies of MST [40]. The study concluded that if therapists adhere to MST, the outcome is better. Low treatment fidelity was a consequence of the failure to provide weekly supervision. The implications of these findings for both policy and practice of MST are clear. Supervision is required if treatment fidelity is to be ensured.
The MST approach has been applied to severe mental health problems of adolescents who require hospitalization [41]. Key findings from this important study were that MST could be offered to acutely disturbed youth as an overall alternative to hospitalization. Multisystemic therapy interventions were able to more effectively reduce externalizing symptoms, reduce hospitalization and improve family cohesion. However, the study also found that MST workers needed much more intense supervision, smaller case loads, required the services of brief hospitalization and the use of medication in over 40% of MST-treated cases.
Recommendations
The adoption of evidence-based practice within the culture of CYMHS is a large topic in its own right and can only be briefly commented on within this review. On one hand there is little to be gained through the imposition of a narrow hegemony of laboratory-based approaches with the associated destruction of established traditions. On the other hand, rejection or denial of the hard-won knowledge base, accumulated over several decades of research is equally destructive.
CYMHS role in promotion, prevention and treatment
Based on the efficacious programs reviewed above, one program stands out as a possible candidate for CYMHS involvement in terms of mental health promotion (the improvement of wellbeing). The training of large numbers of primary school-aged children in the use of social problem-solving as described in the PATHS project could provide a forum for close involvement of CYMHS with primary school systems [33]. While direct involvement of CYMHS staff in the classroom component is beyond its resources, participation may occur with the high-risk groups of children, either through parent groups or social skills groups for children.
Home visiting services based around the principles of David Olds' work provides the best prevention model during the infancy period [30]. Such a project once again would be beyond the resources of CYMHS and are more logically the province of an antenatal and child health service. However, close CYMHS liaison with such services could allow for the development of an infant psychiatry program for high-risk pregnant and post-partum mothers unresponsive to the initial intervention. Child and youth mental health services would have the advantage of multidisciplinary assessment and treatment choices, including a home-visiting capacity for high-risk mothers and infants.
It is during the preschool and primary school periods that both prevention and treatment strategies more closely merge with therapies traditionally associated with CYMHS, for example individual and family approaches. In arguing for the adoption of the below schemes, it is assumed that CYMHS will make the necessary commitments to training and quality control. From the prevention perspective this review has highlighted the place of parent management training and parent education in child development, such as the High/Scope Perry Preschool Project [31] and the ‘Incredible Years’ programs [32]. Within Australia there is an established prevention program that fits the needs of the preschool and early primary school population, the Positive Parenting Program (Triple-P) [42]. A major innovation of Triple-P is that the intervention menu has been conceptualized with five different levels of intensity depending on the nature of the child's problems and parenting risk variables. Level 1 offers self-help information only; Level 2 offers information plus minimal therapist contact; Level 3 involves information plus active skills training; Levels 4 and 5 require increasing therapist expertise in order to deal with more disruptive behaviour and complex parental risk factors such as depression and domestic violence. Once again CYMHS may not be able to administer the whole program but may be best placed to intervene at Level 4 and 5. The training in the use of manualized social problem-solving programs, either PATHS [33] or CSCT [36] should receive a high priority within CYMHS as these approaches, especially the later, have been demonstrated to reduce aggression.
An evidence-based CYMHS response to adolescent aggression/antisocial behaviour has one clear model to follow. Multisystemic therapy has established itself as the outstanding candidate for efficacious intervention in the adolescent age group. It is the adolescent group, especially in the form of severe ODD or complex CD (with comorbid ADHD, language disorders and drug abuse) that present the most serious challenge to CYMHS and other services. Equally this group of adolescents present special problems for the MST model in that there is an increased demand for more clinical resources as well as the need for occasional hospitalization [40]. The use of combined community and hospital mental health facilities raises the need for careful coordination between both sectors when dealing with this complex group. New Zealand now has an established, approved MST program. This new situation has the potential to offer accredited training in this intervention, making such an enterprise both accessible and affordable to Australian mental health service providers. Any strategy to adopt MST in this country must involve careful testing of the model under typical clinical conditions, plus measurement of the savings that hopefully would emerge from this intervention. This could be best done through a combination of national and state government funding initiatives with the support of the services likely to benefit from such an intervention, for example education, child welfare and juvenile justice.
One of the implications of MST research has been the place of evidence-based supervision as a critical element in the success of its program [39]. This finding could have wide-ranging effects on the nature of staff supervision as practised in CYMHS. While a variety of supervision models probably take place in CYMHS, MST research points to a focused, outcome-driven, measurement- based continuous review process for clinicians. Such a supervision model may only work in the context of broad acceptance of a range of evidence-based activities successfully embedded within CYMHS.
Conclusion
Two broad forces are now influencing CYMHS. First, research in the last three decades has produced a broad consensus on the origins, trajectory and prevention/ treatment strategies for aggression/antisocial behaviour in children and adolescents. This knowledge base provides a challenge to CYMHS in terms of integrating these findings into a range of service endeavours from promotion, prevention and treatment. Second, CYMHS will continue to be influenced by the evolving National Mental Health Strategy, especially in terms of the demand for evidence-based practice. Incorporating these demands into CYMHS will be a major challenge. The adoption of the programs outlined will require strengthening of partnerships with home-based nurse visitors and school-based intervention programs. As well, commitment to specific training in parent management and child social cognitive therapy skills is essential. Finally, the extension and reform of CYMHS to enable both infant psychiatry and MST to develop will require the commitment of significant resources to such program development. Despite these difficulties, CYMHS have a responsibility to make an effective contribution to one of Australia's costly community problems: child and adolescent aggression and antisocial behaviour.
Footnotes
Acknowledgements
I thank Barry Nurcombe, former Director, Royal Children's Hospital Health District Child and Youth Mental Health Service and Brett McDermott, Director, Mater Health Services Child and Youth Mental Health Service.