Antisocial behaviour and conduct disorders in children and young people. Costing report. Implementing NICE guidance

Antisocial behaviour and conduct disorders in children and young people Costing report Implementing NICE guidance March 2013 and young people (March 2013) 1 of 44 NICE clinical guideline 158

This costing report accompanies the clinical guideline: Antisocial behaviour and conduct disorders in children and young people: recognition, intervention and management (available online at www.nice.org.uk/guidance/cg158). Issue date: March 2013 The guidance is written in the following context This report represents the view of NICE, which was arrived at after careful consideration of the available data and through consulting with healthcare professionals. It should be read in conjunction with the NICE guideline. The report is an implementation tool and focuses on the recommendations that were considered to have a significant impact on national resource utilisation. Assumptions used in the report are based on assessment of the national average. Local practice may be different from this, and the impact should be estimated locally. Implementation of the guidance is the responsibility of local commissioners and/or providers. Commissioners and providers are reminded that it is their responsibility to implement this guidance, in their local context, in light of their duties to avoid unlawful discrimination and to have regard to promoting equality of opportunity. Nothing in the costing assessment should be interpreted in a way that would be inconsistent with compliance with those duties. National Institute for Health and Clinical Excellence Level 1A City Tower Piccadilly Plaza Manchester M1 4BT www.nice.org.uk National Institute for Health and Clinical Excellence, 2013. All rights reserved. This material may be freely reproduced for educational and not-for-profit purposes. No reproduction by or for commercial organisations, or for commercial purposes, is allowed without the express written permission of NICE. and young people (March 2013) 2 of 44

Contents Executive summary... 4 Potential resource-impact recommendations... 4 Costs... 5 Benefits and savings... 6 1 Introduction... 8 1.1 Supporting implementation... 8 1.2 What is the aim of this report?... 8 1.3 Epidemiology of antisocial behaviour and conduct disorders in children and young people... 8 1.4 Resource impact... 11 1.5 Current service provision... 12 2 Costing methodology... 14 2.1 Process... 14 2.2 Scope of the cost-impact analysis... 14 3 Analysis of the potential resource impact... 15 3.1 Cost of intervention to prevent conduct disorders in children and young people... 15 3.2 Cost of interventions for children and young people with conduct disorders... 16 3.3 Potential savings... 29 4 Impact of guidance for commissioners... 42 5 Conclusion... 42 Appendix A. Approach to costing guidelines... 43 Appendix B. References... 44 and young people (March 2013) 3 of 44

Executive summary This costing report looks at the resource impact of implementing the NICE guideline Antisocial behaviour and conduct disorders in children and young people: recognition, intervention and management in England. Commissioners are encouraged to complete the accompanying costing template to obtain the full benefits of modelling the likely resource impacts for their local area. The costing method adopted is outlined in appendix A; it uses the most accurate data available, was produced in conjunction with key clinicians, and reviewed by clinical and financial professionals. Potential resource-impact recommendations This report focuses on the recommendations that are likely to have the greatest resource impact and therefore require the most additional resources to implement or can potentially generate the biggest savings. They are: Offering classroom-based emotional learning and problem-solving programmes for children aged typically between 3 and 7 years in schools where classroom populations have a high proportion of children identified to be at risk of developing oppositional defiant disorder or conduct disorder because of particular factors. Offering interventions that are suitable for children and young people at high risk of or with oppositional defiant disorder or conduct disorder, and/or their parents, foster carers or guardians. The clinical guideline recommends a number of different types of intervention for children and young people with conduct disorders and/or their parents or carers. These include: parent training programmes foster carer/guardian training programmes parent and child training programmes for children with severe and complex needs and young people (March 2013) 4 of 44

child-focused programmes multimodal interventions. Because of the variation in current practice and uncertainty about future practice, it is not possible to quantify the national cost impact of implementing the guidance. This report discusses the potential costs and savings that need to be considered at a local level. Costs Conduct disorders in children and young people are common. Estimates vary, but the costing tools assume that 5% of children and young people have a conduct disorder (including oppositional defiant disorder). In addition to this defined diagnostic condition, a further group has been categorised within research as having mild to moderate conduct problems and their behaviour also has cost implications. As this group is part of a continuum of behaviour, the size of this cohort is subjective, but studies have looked at the costs associated with a group of between 9-45% of children and young people (Fergusson et al. 2005). Depending on the complexity and severity of the problem, families may receive more than 1 type of intervention. The proportion of people given each type of intervention is likely to vary locally and should be estimated on a local basis. Therefore the total resource impact of implementing the guideline is uncertain. Based on a number of assumptions detailed in section 3, it is estimated that the cost per person of each intervention is as follows: and young people (March 2013) 5 of 44

Intervention Cost per child of providing a child-focused programme Cost per parent or foster carer/guardian of providing a group parent- or foster carer/guardian training programme Cost per parent or foster carer/guardian of providing an individual parent- or foster carer/guardian training programme Cost per family of providing a multimodal intervention Estimated cost per intervention( ) 900 1600 2200 3300 (depending on whether the programme is homeor clinic-based) 15,400 Benefits and savings It is anticipated that providing successful interventions to children and young people with conduct problems and/or their parents or carers will result in savings. Adults who had a conduct disorder or conduct problems in childhood are estimated to be responsible for 80% of crime (Sainsbury Centre for Mental Health 2009). They also incur additional NHS, social services and education costs from childhood (Bonin et al. 2011). Successful intervention will result in a decrease in the prevalence of conduct disorder and conduct problems and reduce costs. Savings will arise across public sectors including the NHS, social services, education and the criminal justice system, as well as private costs such as insurance and replacing stolen property. The lifetime savings are anticipated to be considerably higher than the cost of intervention. The lifetime savings per person will differ depending on the intervention received and the success rates achieved. Based on a number of assumptions detailed in section 3, the potential average lifetime savings to the NHS and social services per person are estimated to range between 4700 and 24,800. Estimated lifetime savings increase to between 23,800 and 104,900 when the education and the criminal justice systems are taken into account. Further details are given in section 3. and young people (March 2013) 6 of 44

Based on a number of assumptions detailed in section 3 and a population of 100,000, it is estimated that there could be potential annual savings as follows: Intervention Child-focused programme Parent training programme, or foster carer/guardian training programme Potential average annual recurrent saving to the NHS and social services ( ) 3000 11,000 1000 13,000 Multimodal intervention 5000 26,000 Local costing template Potential average annual recurrent saving to the NHS, social services, the department of education and the criminal justice system ( ) The costing template produced to support this guideline enables organisations in England, Wales and Northern Ireland to estimate the impact locally and replace variables with ones that depict the current local position. and young people (March 2013) 7 of 44

1 Introduction 1.1 Supporting implementation 1.1.1 The NICE clinical guideline on antisocial behaviour and conduct disorders in children and young people is supported by the following implementation tools available on our website www.nice.org.uk/guidance/cg158. costing tools a national costing report; this document a local costing template; a spreadsheet that can be used to estimate the local cost of implementation tools to help commissioners implement the guidance - a clinical audit tool - a baseline assessment tool 1.2 What is the aim of this report? 1.2.1 This report aims to help organisations plan for the financial implications of implementing NICE guidance. 1.2.2 This report does not reproduce the NICE guideline on antisocial behaviour and conduct disorders in children and young people and should be read in conjunction with it (see www.nice.org.uk/guidance/cg158). 1.2.3 The costing template that accompanies this report is designed to help those assessing the resource impact at a local level in England, Wales or Northern Ireland. 1.3 Epidemiology of antisocial behaviour and conduct disorders in children and young people 1.3.1 Antisocial behaviours and conduct disorders are the commonest mental health disorders of childhood and adolescence globally, and the most common reason for referral to child and adolescent and young people (March 2013) 8 of 44

mental health services in western countries (Antisocial behaviour and conduct disorders in children and young people: recognition, intervention and management). 1.3.2 Disorders in adolescence are becoming more frequent in western countries and can place a large personal and economic burden on individuals and society, involving not just healthcare services and social care agencies but all sectors of society including the family, schools, police and criminal justice agencies (Antisocial behaviour and conduct disorders in children and young people: recognition, intervention and management). 1.3.3 A report by the Office for National Statistics (2005), Mental health of children and young people in Great Britain, 2004, estimates that 6.1% of children and young people aged 5 to 16 years have a clinically diagnosed conduct disorder in England. Figure 1 Prevalence of conduct disorders in children and young people in England 1.3.4 Figure 2 shows how prevalence of conduct disorders varies by ethnicity, with a higher proportion of white and black children being affected by conduct disorders. and young people (March 2013) 9 of 44

Figure 2 Prevalence of conduct disorders by ethnicity 1.3.5 Figure 3 shows a general decrease in conduct disorders as household income increases. Figure 3 Prevalence of conduct disorders by household income 1.3.6 Of those with an early onset conduct disorder (before age 8) about half have serious problems that persist into adulthood. Of those with adolescent onset, the great majority (over 85%) no longer have a conduct disorder (antisocial behaviour) by their early 20s and young people (March 2013) 10 of 44

(Antisocial behaviour and conduct disorders in children and young people: recognition, intervention and management). 1.4 Resource impact 1.4.1 Although conduct disorders can be considered to be primarily a mental health problem (DSM-IV, American Psychiatric Association 2000), healthcare service provision for conduct disorders and healthcare costs are small when compared with the costs incurred by sectors such as the criminal justice system (Scott et al. 2001). Crimes committed as a result of conduct disorders lead to significant social costs and harm to individuals and their victims, families and carers, and society at large (Welsh et al. 2008). Overall, evidence on the costs arising from conduct disorders varies widely and is greater when a societal perspective is taken (Antisocial behaviour and conduct disorders in children and young people: recognition, intervention and management). 1.4.2 In a UK study by Scott et al. (2001), the cumulative cost of services used by children and young people diagnosed with conduct disorder at the age of 10 years was 70,000 (1998 costs) over a period of around 18 years. Costs incurred by children and young people with conduct disorders are about 10 times more than those incurred by people with no conduct problem. 1.4.3 In a UK study by Romeo et al. (2006), the annual cost per child with conduct disorder (excluding criminal justice costs) was estimated to be about 5960 (2002/03 costs), with costs to the family accounting for about 79% of the total cost and health services, education and voluntary services costs accounting for about 8%, 1% and 3% respectively. The cost to social services was estimated to be less than 1% of the total cost. Similarly, a study by Knapp et al. (1999) estimated the annual mean cost of services for 10 children aged 4 to 10 years as being 15,270 (1996/97 prices), with costs to their and young people (March 2013) 11 of 44

families accounting for about 31% and costs to health services accounting for 16%. 1.4.4 A report by the Sainsbury Centre for Mental Health (2009) estimates that around 80% of all criminal activity is attributable to people who had conduct problems in childhood and adolescence. The costs to society are substantial. For example, the lifetime costs of crime committed by a single prolific offender are about 1.5 million. The total cost of crime attributable to people who had conduct problems in childhood is estimated to be about 60 billion a year in England and Wales. 1.4.5 The Sainsbury Centre for Mental Health estimates that the overall lifetime cost of adverse outcomes among the 5% of people who have conduct disorder in childhood is around 225,000 per person. The lifetime cost among those who have mild or moderate conduct problems in childhood is around 75,000 per person (based on a 45% of children and young people). Crime-related costs make up 71% of the estimated total lifetime cost for people with conduct disorder and 61% for people with a mild or moderate conduct problem. 1.5 Current service provision 1.5.1 Conduct disorders are the most common reason for referral of young children to child and adolescent mental health services (CAMHS). Children with conduct disorders also account for much of the work of the health and social care system. For example, 30% of a typical GP s child consultations are for conduct disorders, 45% of community child health referrals are for behaviour disturbances, and psychiatric disorders are a factor in 28% of all paediatric outpatient referrals. In addition, social care services have significant involvement with children and young people with conduct disorders, with more vulnerable or disturbed children often being placed with foster families or, in a small number of cases, in and young people (March 2013) 12 of 44

residential care. The demands on the education system are also considerable and include the provision of special-needs education. The criminal justice system has significant involvement with older children with conduct disorders (Antisocial behaviour and conduct disorders in children and young people: final scope). 1.5.2 Multiple agencies may be involved in the care and treatment of children with conduct disorders, which presents a major challenge for the effective coordination of care (Antisocial behaviour and conduct disorders in children and young people: final scope). 1.5.3 Several interventions have been developed for children with conduct disorders and related problems, such as parent training programmes. Other interventions focused on prevention, such as the Family Nurse Partnership Programme, have recently been implemented in the UK and are being evaluated. Three themes are common to these interventions: a strong focus on working with parents and families, recognition of the importance of the wider social system in enabling effective interventions, and a focus on preventing or reducing the escalation of existing problems (Antisocial behaviour and conduct disorders in children and young people: final scope). 1.5.4 Uptake of these interventions varies across the country. Parenting programmes are the best established; implementation of multisystemic approaches and early intervention programmes is more variable. In addition to the programmes developed specifically for children and young people with conduct disorders, treatments are offered by both specialist CAMHS teams and general community-based services such as Sure Start (Antisocial behaviour and conduct disorders in children and young people: final scope). 1.5.5 At present less than 25% of affected children and young people receive any specific help (Vostanis et al. 2003), and much of the help provided is likely to be ineffective (Scott 2007). and young people (March 2013) 13 of 44

2 Costing methodology 2.1 Process 2.1.1 We use a structured approach for costing clinical guidelines (see appendix A). 2.1.2 We have to make assumptions in the costing model. These are tested for reasonableness with members of the Guideline Development Group and key clinical practitioners in the NHS. 2.2 Scope of the cost-impact analysis 2.2.1 The guideline offers best practice advice on antisocial behaviour and conduct disorders in children and young people. 2.2.2 The guidance covers: Children and young people (aged 18 years and younger) with a diagnosed or suspected conduct disorder, including looked after children and those in contact with the criminal justice system. Children and young people identified as being at significant risk of developing conduct disorders. 2.2.3 The guidance does not cover: Adults (aged 19 and older). Children and young people with coexisting conditions if conduct disorder is not a primary diagnosis. Children and young people with psychosis. Children and young people with autism spectrum conditions. Primary drug and alcohol problems. Children and young people with speech and language difficulties whose behavioural problems arise from the speech and language difficulties. Therefore, these issues are outside the scope of the costing work. and young people (March 2013) 14 of 44

2.2.4 Rather than cost each individual recommendation, costing work has focused on the areas that will potentially need the most resources to implement or generate the biggest savings. These areas were determined in discussion with the clinical guideline project team and the members of the Guideline Development Group. 3 Analysis of the potential resource impact 3.1 Cost of intervention to prevent conduct disorders in children and young people Recommendations Offer classroom-based emotional learning and problem-solving programmes for children aged typically between 3 and 7 years in schools where classroom populations have a high proportion of children identified to be at risk of developing oppositional defiant disorder or conduct disorder as a result any of the following factors: low socioeconomic status low school achievement child abuse or parental conflict separated or divorced parents parental mental health or substance misuse problems parental contact with the criminal justice system. Classroom-based emotional learning and problem-solving programmes should be provided in a positive atmosphere and consist of interventions intended to: increase children s awareness of their own and others' emotions teach self-control of arousal and behaviour promote a positive self-concept and good peer relations develop children s problem-solving skills. and young people (March 2013) 15 of 44

Typically the programmes should consist of up to 30 classroom-based sessions over the course of 1 school year. Background 3.1.1 According to a report by the Sainsbury Centre for Mental Health Potential costs (2009), programmes aimed at prevention or early intervention at pre-school age are the most effective for avoiding childhood conduct problems. 3.1.2 Classroom-based interventions are likely to vary locally, and may or may not be teacher-led. The cost of a representative intervention, including teacher training, programme coordinator and materials, has been taken from Curtis (2011) and is 132 per child per year (2009 prices). Uplifting prices for 2011 using the Hospital and Community Health Services (HCHS) index gives an estimated cost of 136. 3.1.3 It is not known how many children aged between 3 and 7 years would be assessed as being at risk of developing oppositional defiant disorder or conduct disorder. Please estimate the cost locally. 3.2 Cost of interventions for children and young people Background with conduct disorders 3.2.1 The clinical guideline recommends a number of different types of intervention for children and young people with conduct disorders. These include: parent training programmes foster carer/guardian training programmes parent and child training programmes for children with severe and complex needs and young people (March 2013) 16 of 44

child-focused programmes multimodal interventions. 3.2.2 According to clinical opinion, there are no validated evidence-based child-focused interventions and programmes, and access to programmes is likely to vary locally. Child-focused programmes are for the child only and most interventions for children with conduct disorders are based on behavioural or cognitive behavioural principles (Antisocial behaviour and conduct disorders in children and young people: recognition, intervention and management). 3.2.3 According to clinical opinion, parent training programmes follow set structures, and are based on licensed programmes such as the Webster-Stratton Incredible Years programme. However, access to parent training programmes varies, and about half the population may not have access. 3.2.4 The Incredible Years programme is based on social learning theory and consists of weekly 2-hour group sessions over 12 weeks, delivered by a trained practitioner (Ministry of Justice: the Incredible Years factsheet). 3.2.5 If children have severe and complex problems, the clinical guideline recommends that individual parent and child training programmes are offered. 3.2.6 Multimodal interventions are the most intensive and expensive interventions and are recommended for children and young people with conduct disorders aged 11 17 years. Multimodal interventions involve both the child or young person and their family. People who receive a multimodal intervention would not be offered any of the other interventions. 3.2.7 The costing report and template use multisystemic therapy (MST) as an example of a multimodal intervention. Multisystemic therapy (MST) is a type of multimodal intervention that was specifically and young people (March 2013) 17 of 44

developed for working with adolescents with conduct disorders (Antisocial behaviour and conduct disorders in children and young people: recognition, intervention and management). 3.2.8 According to clinical opinion, MST programmes follow set structures. MST is a licensed and evidence-based community intervention for children and young people aged 11 17 years and their families. 3.2.9 MST is supported by the Department for Education, the Youth Justice Board and the Department of Health, and has been trialled across 10 sites in England. The 10 sites are: Barnsley; the London boroughs of Greenwich, Hackney, and Merton and Kingston; Leeds; Peterborough; Plymouth; Reading; Sheffield and Trafford. All sites are partnerships between the local authority/children's trust (including CAMHS), the primary care trust and the Youth Offending Service and also collaborate with adult mental health and substance misuse services. 3.2.10 According to clinical opinion most of these teams are still in place and there are currently around 15 20 teams providing MST in the UK. Teams tend to cover only small areas and so most localities do not currently have access to MST teams. 3.2.11 Where MST programmes are available, they have been completed by 90% of the families taking part, and 89% of young people still live at home by the end of the programme (National mental health development unit). General assumptions 3.2.12 Estimates of the number of children and young people with conduct disorders vary. The costing template assumes that 5% of children and young people have conduct disorder (Fergusson et al. 2005). and young people (March 2013) 18 of 44

3.2.13 Depending on the complexity and severity of the problem, families may receive more than one type of intervention. The proportion of people given each type of intervention is likely to vary locally and should be estimated on a local basis. Therefore the resource impact of implementing the guideline is uncertain. The costing template that accompanies this report can be used as a basis for calculating the local resource impact based on local circumstances. 3.2.14 It is assumed that there will be an initial cost of providing an intervention to the current eligible population, the cost of which may be staggered over a number of years depending on how quickly the programmes are rolled out, and an annual recurrent cost of providing an intervention programme based on the average population becoming eligible each year (although children may develop symptoms at any point within the eligible range). 3.3 Parent training programmes Recommendations Offer a group parent training programme to the parents of children and young people aged between 3 and 11 years who: have been identified as being at high risk of developing oppositional defiant disorder or conduct disorder or have oppositional defiant disorder or conduct disorder or are in contact with the criminal justice system because of antisocial behaviour. Group parent training programmes should involve both parents if this is possible and in the best interests of the child or young person, and should: typically have between 10 and 12 parents in a group be based on a social learning model, using modelling, rehearsal and feedback to improve parenting skills typically consist of 10 to 16 meetings of 90 to 120 minutes' duration and young people (March 2013) 19 of 44

adhere to the developer s manual (as used in relevant randomised controlled trials) and employ all of the necessary materials to ensure consistent implementation of the programme. Offer an individual parent training programme to the parents of children and young people aged between 3 and 11 years who are not able to participate in a group parent training programme and whose child: has been identified as being at high risk of developing oppositional defiant disorder or conduct disorder or has oppositional defiant disorder or conduct disorder or is in contact with the criminal justice system because of antisocial behaviour. Individual parent training programmes should involve both parents if this is possible and in the best interests of the child or young person, and should: be based on a social learning model using modelling, rehearsal and feedback to improve parenting skills typically consist of 8 to 10 meetings of 60 to 90 minutes duration adhere to the developer s manual (as used in relevant randomised controlled trials) and employ all of the necessary materials to ensure consistent implementation of the programme. Potential costs 3.3.1 Based on expert opinion, it is estimated that 50% of parents offered a parent training programme would take up the offer. Of these, 20% would be unable to engage in group parent training programmes and be offered an individual programme. It is estimated that 50% of individual programmes would be clinic-based and 50% would be home-based. The cost of a clinic-based individual parent training programme is estimated to be 2210 and the cost of a home-based individual parent training programme is estimated to be 3315. Training is estimated to cost 1133 per facilitator, with 5 individual programmes run by a single facilitator per year. (The costs for and young people (March 2013) 20 of 44

2006/07 have been uplifted to reflect 2011 prices using the HCHS index; Curtis 2011). 3.3.2 Based on the Webster-Stratton Incredible Years programme, a group parent training programme based on 12 parents per group is estimated to cost 1209 per parent excluding set-up costs or 1600 including set-up costs (Curtis 2011). 3.3.3 For a population of 100,000, there would be about 500 children aged between 3 and 11 years with conduct disorder, and about 60 children a year who develop it. There would also be about 4700 with a conduct problem, and 500 children who develop it each year. 3.3.4 It is not known what proportion of children and young people with conduct disorders are cared for by their parents rather than foster carers or guardians. The incremental number of parent-focused programmes that would be provided is uncertain. 3.3.5 Based on the assumptions above and a population of 100,000, figure 4 illustrates how the cost increases as the number of parentfocused programmes provided increases. Figure 4 Illustration of the increase in costs per 100,000 population and young people (March 2013) 21 of 44

3.4 Foster carer/guardian training programmes Recommendations Offer a group foster carer/guardian training programme to foster carers and guardians of children and young people aged between 3 and 11 years who: have been identified as being at high risk of developing oppositional defiant disorder or conduct disorder or have oppositional defiant disorder or conduct disorder or are in contact with the criminal justice system because of antisocial behaviour. Group foster carer/guardian training programmes should involve both of the foster carers or guardians if this possible and in the best interests of the child or young person, and should: modify the intervention to take account of the care setting in which the child is living typically have between 8 and 12 foster carers or guardians in a group be based on a social learning model using modelling, rehearsal and feedback to improve parenting skills typically consist of between 12 and 16 meetings of 90 to 120 minutes duration adhere to the developer s manual (as used in relevant randomised controlled trials) and employ all of the necessary materials to ensure consistent implementation of the programme. Offer an individual foster carer/guardian training programme to the foster carers or guardians of children and young people aged between 3 and 11 years at high risk of or with oppositional defiant disorder or conduct disorder who are not able to participate in a group programme and whose child: has been identified as being at high risk of developing oppositional defiant disorder or conduct disorder or has oppositional defiant disorder or conduct disorder or and young people (March 2013) 22 of 44

is in contact with the criminal justice system because of antisocial behaviour. Individual foster carer/guardian training programmes should involve both of the foster carers if possible and in the best interests of the child or young person, and should: modify the intervention to take account of the care setting in which the child is living be based on a social learning model using modelling, rehearsal and feedback to improve parenting skills consist of up to 10 meetings of 60 minutes duration adhere to the developer s manual (as used in relevant randomised controlled trials) and employ all of the necessary materials to ensure consistent implementation of the programme. Potential costs 3.4.1 It is not known what proportion of children and young people with conduct disorders are cared for by foster carers or guardians as opposed to their parents. Because current and future practice are uncertain, it is not possible to estimate the incremental number of foster carer/guardian-focused programmes that would be provided. 3.4.2 The criteria for foster carer/guardian-focused programmes is very similar to the parent-focused programme. Therefore it is anticipated that the cost of the Incredible Years parenting programme can be used as a proxy, along with the assumptions made for the parentfocused programme. Please enter local estimates on the costing template. and young people (March 2013) 23 of 44

3.5 Child-focused interventions for children and young people with conduct disorders Recommendations Offer group social and cognitive problem-solving programmes to children and young people aged between 9 and 14 years who: have been identified as being at high risk of developing oppositional defiant disorder or conduct disorder or have oppositional defiant disorder or conduct disorder or are in contact with the criminal justice system because of antisocial behaviour. Group social and cognitive problem-solving programmes should be adapted to the children s or young people's developmental level and should: be based on a cognitive behavioural problem-solving model use modelling, rehearsal and feedback to improve skills typically consist of 10 to 18 weekly meetings of 2 hours duration adhere to the developer s manual (as used in relevant randomised controlled trials) and employ all of the necessary materials to ensure consistent implementation of the programme. Potential costs 3.5.1 According to clinical opinion, programmes are carried out by individual clinicians such as clinical psychologists, community nurses or social workers with relevant training and experience. Staff taking the sessions are typically agenda for change band 7, with some supervision by a band 8a senior member of staff. 3.5.2 The cost of providing the programme will differ locally depending on: the increase in the number of children and young people offered a child-focused programme, uptake, the number of weekly meetings, the staff used to run the programme and the number of children and young people per group. and young people (March 2013) 24 of 44

3.5.3 Based on the assumptions used in the health economic modelling, it is estimated that a child-focused programme could cost around 890 per person (please see the costing template that accompanies this report for more details on the assumptions used and to tailor the cost based on local estimates). 3.5.4 As both current and future practice are uncertain, it is not possible to estimate the incremental number of child-focused programmes that would be provided. 3.5.5 For a population of 100,000, there would be about 240 children aged between 7 and 14 years with conduct disorder, and about 30 a year who develop it. There would also be around 2100 with conduct problems and 260 who develop it each year. 3.5.6 Based on a population of 100,000, and an assumed uptake rate of 50%, figure 5 illustrates how the cost increases as the number of child-focused programmes provided increases. and young people (March 2013) 25 of 44

Figure 5 Illustration of the increase in costs per 100,000 population 3.6 Parent and child training programmes for children with complex needs Recommendations Offer individual parent and child training programmes to children and young people aged between 3 and 11 years if their problems are severe and complex and they: have been identified as being at high risk of developing oppositional defiant disorder or conduct disorder or have oppositional defiant disorder or conduct disorder or are in contact with the criminal justice system because of antisocial behaviour. Individual parent and child training programmes should involve both parents, foster carers or guardians if this is possible and in the best interests of the child or young person, and should: be based on a social learning model using modelling, rehearsal and feedback to improve parenting skills consist of up to 10 meetings of 60 minutes duration and young people (March 2013) 26 of 44

adhere to the developer s manual (as used in relevant randomised controlled trials) and employ all of the necessary materials to ensure consistent implementation of the programme. Potential costs 3.6.1 If children have severe and complex needs, the clinical guideline recommends that parents receive a parenting programme and children receive a child-focused programme. Alternatively, for children and young people with severe conduct disorder and complex needs in the eligible age range, the whole family may receive a multimodal intervention such as multisystemic therapy. 3.6.2 It is not known what proportion of children and young people with conduct disorders have severe and complex needs. Please use the costing template that accompanies this report to estimate the cost locally. 3.7 Multimodal interventions (based on multisystemic therapy in the costing model) Recommendations Offer multimodal interventions (for example, multisystemic therapy) to children and young people aged between 11 and 17 years for the treatment of conduct disorder. Multimodal interventions should involve the child or young person and their parents and carers and should: have an explicit and supportive family focus be based on a social learning model with interventions provided at individual, family, school, criminal justice and community levels be provided by specially trained case managers typically consist of 3 to 4 meetings per week over a 3- to 5-month period and young people (March 2013) 27 of 44

adhere to the developer s manual (as used in relevant randomised controlled trials) and employ all of the necessary materials to ensure consistent implementation of the programme. Potential costs 3.7.1 Based on the assumptions used in the health economic modelling, it is estimated that a multisystemic therapy could cost around 15,400 per family. (Please see the costing template that accompanies this report for more details on the assumptions used and to tailor the cost based on local estimates.) 3.7.2 Therapists will need to attend an initial training course and 4 booster training sessions a year. The initial training cost is shown separately on the costing template as a non-recurrent cost. The health economic modelling estimates that the initial training cost per therapist would be 6000. If the impact of training is estimated to last 5 years, in which time a therapist carried out 45 interventions, the initial training cost per intervention is estimated to be 133. 3.7.3 The health economic model assumes that treatment would usually be with a youth offending team, the cost of which is estimated to be around 2300. If multimodal interventions were to replace working with youth offending teams then cost savings to the Ministry of Justice would be likely. 3.7.4 As both current and future practice are uncertain, it is not possible to estimate the incremental number of multimodal interventions that would be provided. 3.7.5 For a population of 100,000, there would be about 430 children and young people aged between 11 and 17 years with conduct disorder, and about 60 a year who develop it. 3.7.6 Based on a population of 100,000, and an assumed uptake rate of 90% based on the Department of Health trials, figure 6 illustrates and young people (March 2013) 28 of 44

how the cost increases as the number of multimodal interventions provided increases. Figure 6 Illustration of the increase in costs per 100,000 population Potential savings 3.7.7 It is anticipated that providing successful programmes to children and young people with conduct disorder and/or their parents or carers will result in savings. 3.7.8 Conduct disorders can persist into adulthood, at which point they are classified as antisocial personality disorder. Adults who had a conduct disorder or conduct problems in childhood are estimated to be responsible for 80% of crime (Sainsbury Centre for Mental Health 2009). They also incur additional NHS, social services and education costs from childhood (Bonin et al. 2011). 3.7.9 Successful intervention will result in a decrease in the prevalence of conduct disorder and conduct problems and save costs. 3.7.10 Savings will arise across public sectors including the NHS, social services, education and the criminal justice system, as well as savings in private costs such as insurance and replacing stolen property. and young people (March 2013) 29 of 44

3.7.11 The costing template provides separate estimates of the potential savings for the NHS and social services, and the wider public sector including education and crime. 3.7.12 Children and young people with a conduct disorder (including oppositional defiant disorder) will move to having conduct problems or no conduct problems, and children and young people with conduct problems will move to having no conduct problem. 3.7.13 The estimated annual NHS, social services and education costs for people with persistent conduct disorder have been taken from Bonin et al. (2011). This study is based on 2008/09 prices, therefore costs have been uplifted to estimate 2011 prices using the HCHS index (Curtis 2011). The costs are summarised in table 1. Table 1 Estimated annual NHS, social services and education costs per person for people with conduct disorder (Bonin et al. 2011, uplifted for 2011 prices) Department Age 5 10 Age 11 16 Age 17+ NHS and social services 1313 170 170 Education 912 1243 0 3.7.14 Scott et al. (2001) has been used to estimate the relative costs incurred for people with conduct problems or no conduct problem compared with people who have a conduct disorder. Table 3 in Scott et al. (2001) estimates the arithmetic mean of total costs (in at 1998 prices) of services used by people up to the age of 28 (according to group at age 10) as 7423 for people with no conduct problem, 24,324 for people with conduct problems and 70,019 for people with conduct disorder. This equates to a ratio of 0.11:0.35:1. The relative costs for people with conduct problems or no conduct problem are summarised in table 2. and young people (March 2013) 30 of 44

Table 2 Estimated annual NHS, social services and education costs per person for people with conduct problems or no conduct problem (Scott et al. 2001, uplifted for 2011 prices) Age 5 10 Condition Proportion of estimated annual costs for persistent conduct disorder NHS and social services Conduct disorder 100% 1313 912 Conduct problems 35% 459 319 No conduct problems 11% 144 100 Age 11 16 Department of Education Conduct disorder 100% 170 1243 Conduct problems 35% 59 435 No conduct problems 11% 19 137 Age 17+ Conduct disorder 100% 170 0 Conduct problems 35% 59 0 No conduct problems 11% 19 0 3.7.15 The estimated annual costs of crime to the NHS and the criminal justice system have been taken from Brand and Price (2000). This study is based on 1999/2000 prices, therefore costs have been uplifted to estimate 2011 prices using the HCHS index (Curtis 2011). The study covers both England and Wales so costs have been multiplied by 94% to estimate the costs for England only (table 3). and young people (March 2013) 31 of 44

Table 3 Estimated annual cost of crime to the NHS and the criminal justice system Annual cost (million) Estimated cost to health services in England as a consequence of crime Estimated cost to the criminal justice system (including police and victim services) in England in response to crime 1788 12,409 3.7.16 A report by the Sainsbury Centre for Mental Health (2009) was used to estimate the proportion of annual crime attributable to people with conduct disorder, conduct problems and no conduct problem (after adjusting for other risk factors for offending, notably socioeconomic background and cognitive ability). The estimated proportion of the annual cost of crime to the NHS and the criminal justice system attributable to people with conduct disorder, conduct problems or no conduct problem in childhood is summarised in table 4. Table 4 Estimated proportion of the annual cost of crime to the NHS and the criminal justice system attributable to people with conduct disorder, conduct problems or no conduct problem in childhood Conduct disorder Conduct problems No conduct problems Proportion of annual crime attributable to each group (adjusted for other risk factors for offending, notably socioeconomic background and cognitive ability) Annual cost to the criminal justice system attributable to each group (million) 22% 2693 388 58% 7197 1037 20% 2519 363 Annual cost to the NHS attributable to each group (million) 3.7.17 Fergusson et al. (2005) was used to estimate the number of people in England who have conduct disorder, conduct problems or no conduct problem in childhood. This was then used to calculate the annual cost of crime per person with conduct disorder, conduct problems or no conduct problem. The estimated proportion of the and young people (March 2013) 32 of 44

annual cost of crime to the NHS and the criminal justice system attributable to people with conduct disorder, conduct problems or no conduct problem in childhood is summarised in table 5. Table 5 Estimated proportion of the annual cost of crime to the NHS and the criminal justice system attributable to people with conduct disorder, conduct problems or no conduct problem in childhood Conduct disorder Conduct problems No conduct problem Estimated proportion of people with conduct disorder, conduct problems or no conduct problem in childhood Estimated number of people in England in each group (total population in England is assumed to be 51,573,132) Cost to the criminal justice system per person in each group 5% 2,578,657 1044 150 45% 23,207,909 310 45 50% 25,786,566 98 14 Cost to the NHS per person in each group 3.7.18 The estimated annual NHS, social services, education and crime costs saved for people who receive a successful intervention are summarised in table 6. and young people (March 2013) 33 of 44

Table 6 Estimated annual NHS, social services, education and crime costs saved for people who receive a successful intervention Department People with conduct disorder People with conduct problems People with no conduct problem Per person aged 5 10 NHS and social services 1313 459 144 Education 912 319 100 Crime costs to the NHS Crime costs to the criminal justice system Per person aged 11 16 NHS and social services 0 0 0 0 0 0 170 59 19 Education 1243 435 137 Crime costs to the NHS Crime costs to the criminal justice system Per person aged 17+ NHS and social services 150 45 14 1044 310 98 170 59 19 Education 0 0 0 Crime costs to the NHS Crime costs to the criminal justice system 150 45 14 1044 310 98 3.7.19 These can then be used to calculate the decrease in cost for people moving from conduct disorder to conduct problems or no conduct problem and from conduct problems to no conduct problem as a result of receiving a successful intervention. The estimated annual NHS, social services, education and crime costs saved for people who receive a successful intervention are summarised in table 7. and young people (March 2013) 34 of 44

Table 7 Estimated annual decrease in cost per person who receives a successful intervention Department Decrease in cost for people moving from conduct disorder to conduct problem Decrease in cost for people moving from conduct disorder to no conduct problem Decrease in cost for people moving from conduct problem to no conduct problem Per person aged 5 10 NHS and social services NHS, social services, education and the criminal justice system Per person aged 11 16 NHS and social services NHS, social services, education and the criminal justice system Per person aged 17+ NHS and social services NHS, social services, education and the criminal justice system 853 1168 315 1446 1980 534 216 287 71 1758 2340 582 216 287 71 950 1234 284 3.7.20 The health economic modelling for the clinical guideline estimates the success rates of the different interventions. From these estimates the decrease in people with conduct disorder and the proportion of people who move to having either conduct problems or no conduct problem can be estimated. For simplicity, the costing template assumes that where interventions are also recommended for people with oppositional defiant disorder, the decrease in people with conduct problems will be the same as for people with conduct disorder. The estimated reduction in the number of children and young people with conduct disorder or conduct problems as a result of successful intervention is summarised in table 8. and young people (March 2013) 35 of 44

Table 8 Estimated reduction in the number of children and young people with conduct disorder or conduct problems as a result of successful intervention Intervention Estimated decrease in people with conduct disorder Proportion who move to having mild to moderate conduct problems Proportion who move to having no conduct problem Estimated decrease in people with a mild to moderate conduct problems Child-focused intervention Parent or foster carer/guardianfocused intervention 51% 57% 53% 35% 46% 25% 65% 54% 75% 51% 57% N/A 3.7.21 The potential savings are estimated for the NHS and social Multimodal interventions (multisystemic therapy) services only and for the wider public sector including education and the criminal justice system. Savings are also split between initial savings if interventions are provided for everyone currently eligible, and the average recurrent annual saving as interventions are provided to those who become eligible each year. 3.7.22 For simplicity, to calculate the initial savings the average of the saving per person at the start and end-points of the eligible age range is used. The average recurrent annual savings are based on the saving per person at the start-point of the eligible age range. 3.7.23 The estimated potential saving per person depends on their age when they receive the intervention. The relevant number of years in each age band is then applied to the estimated costs in table 7. 3.7.24 The health economic modelling estimates that 50% of people will have a relapse and it is assumed that people will incur costs at a similar rate beyond the limit of 17 years. For simplicity, no savings have been included for people who have a relapse. 3.7.25 For example, child-focused programmes are recommended for children and young people aged between 7 and 14 years with and young people (March 2013) 36 of 44