Condition category
Mental and Behavioural Disorders
Date applied
23/03/2010
Date assigned
13/05/2010
Last edited
28/08/2014
Prospective/Retrospective
Retrospectively registered
Overall trial status
Completed
Recruitment status
No longer recruiting

Plain English Summary

Not provided at time of registration

Trial website

http://www.iop.kcl.ac.uk/departments/?locator=1119

Contact information

Type

Scientific

Primary contact

Dr Sajid Humayun

ORCID ID

Contact details

PO85 De Crespigny Park
Institute of Psychiatry
London
SE5 8AF
United Kingdom
sajid.humayun@kcl.ac.uk

Additional identifiers

EudraCT number

ClinicalTrials.gov number

Protocol/serial number

N/A

Study information

Scientific title

Study of Adolescents' Family Experiences: a randomised controlled trial of functional family therapy with adolescent offenders and antisocial youth

Acronym

SAFE

Study hypothesis

There are a number of factors that may contribute to the likelihood of a young person offending. Key amongst these are family circumstances and styles of parenting, which have been shown to be strongly correlated with delinquency and anti-social behaviour. As a result, the last thirty years has seen an increasing interest in family interventions designed to reduce youth offending, with a number of interventions having been shown to be effective, both in terms of reductions in recidivism and in terms of cost to the taxpayer. However, there is currently a large gap in the UK experience and knowledge about programmes for teenagers. A particularly promising intervention is Functional Family Therapy (FFT) which has strong evidence from the US, and is less intensive and expensive than alternative interventions such as multi-systemic therapy and multidimensional treatment foster care. As such, FFT may be the most suitable candidate for widespread utilisation by statutory service providers such as Youth Offending Teams (YOTs).

FFT aims to assist young people and their families to make meaningful changes in their functioning. This is accomplished by reducing negativity in the family and focusing on significant yet obtainable behavioural changes that will have a lasting impact on family relationships. These changes may have the immediate effect of changing a specific functional problem or difficulty and the longer term effect of empowering a family to continue to apply changes to future circumstances. A key duty of the therapist is to assume some of the functional duties of a "family case manager" to assist the family to identify and interact with relevant community resources that will help the family sustain their changes over time. In this way, FFT aims to increase the family's capacity to manage problems, enabling a reduced reliance on service providers and their resources.

FFT is supported by over 30 years of clinical research, the results of which indicate that when the intervention is properly implemented the program can successfully reduce recidivism rates, reduce the time spent by juvenile offenders in institutions and reduce the likelihood of delinquency. The implications of the costs on service provision are evidenced by the fact that FFT dramatically reduces the cost of treatment related to juvenile offending. FFT has also demonstrated significant reductions in potential new offending for siblings of the target of an FFT intervention. Both the reduction in resource use for an identified at-risk young person and their siblings provide support for the argument that successful completion of FFT may reduce future reliance on intervention and community services.

Two issues limit the evidence for FFT as an intervention ready to be disseminated into existing service settings. First, existing trials have not compared FFT to an equivalent dose control group and thus it is unknown if the intervention is producing its effects from specific family changes or simply due to extra attention being given to families. Second, given there have not been any randomised controlled trials of FFT conducted in this country, it is not known whether, and to what extent, the intervention can be effectively implemented in UK service agencies and improve outcomes for UK adolescents and their families. Thus, an evaluation of the intervention is likely to have implications for a wider national dissemination of FFT and its use by YOTs and other agencies. In addition, it may also impact on government policy relating to solutions to youth offending and the relative cost-benefits of family therapies in comparison to detention and other measures used by the criminal justice system.

Whilst the evidence of the effect of FFT on recidivism in US populations is compelling it is not clear what mediates this effect. Although the intervention aims to reduce youth re-offending by improving family functioning, this process has not been studied systematically. Improvements in child behaviour may be mediated through changes in a child's attitudes toward parental discipline. Alternatively, changes in parental behaviour or attributions may mediate changes in child behaviour. Lack of knowledge about the mechanisms responsible for change limit the extent to which the programme can be adapted to local settings without compromising its effectiveness.

Therefore, in addition to evaluating the effectiveness of the intervention programmes, the specific mediating and moderating factors will be tested. These will include factors which can be loosely grouped into adolescent, family, and therapy delivery variables. In terms of adolescent variables, we will characterise offenders in terms of sociodemographics, offending history, psychological health and disorder profile, and biological vulnerability. Family variables will include structure and function and relationship quality to the target youth. Service provision variables will include the youth and families engagement with the intervention as well as therapist skill and delivery.

As part of the adolescent variables, we will evaluate a range of biomarkers that have been shown to influence the course of antisocial behaviour. These will range from simple resting arousal levels (heart rate) to specific candidate genes that appear to moderate the effects of environment on antisocial behaviour and violence. Involvement in an intervention programme designed to change parenting and ameliorate adolescent behaviour problems constitute major environmental experiences, and it is likely that the response to such an experience is at least in part genetically mediated.

Over the last decade, gene-environment interplay in many aspects of child and adult psychiatry and psychology has increasingly emerged as one of the most critical for human development. In particular, there have been major advances in understanding the complex relationships between specific genetic variants and environments - including family relationships and parenting - in the development of antisocial behaviour. Despite the pace of developments in this area, there is still much to learn, particularly about how these gene-environment interactions relate to treatment efficacy. The exploration of these important issues within intervention studies is severely limited and the proposed study will thus advance our understanding of genetic sensitivity to behavioural interventions. In turn our findings have the potential to inform future provision for at risk groups, and, similarly, to benefit developers of intervention programmes in their understanding of what works for whom, and why.

Thus, the study will investigate the direct and indirect role of a number of genetic variants in the response to treatment. These will include MAOA, 5-HTT, and COMT polymorphisms all of which have been shown to moderate environmental influences of antisocial behaviour and psychopathology. For example, there is evidence that genetic deficiencies in MAOA activity are linked to aggression, and maltreatment is known to have links with antisocial behaviour; recent findings have suggested an interaction between these genetic and environmental influences, in that maltreated children with high levels of MAOA expression may be less likely to develop antisocial behaviour. On the basis of these and other similar research findings we propose an investigation of MAOA and other candidate genes (e.g., COMT, 5HT) in interaction both with parenting and with intervention. We will use a standard method for collecting DNA samples that is used in numerous studies, including several based at King's College London. That is, we will collect cheek swabs from participants.

In summary, the current study will conduct evaluation of FFT to reduce adolescent offending behaviour within existing UK statutory service provider settings. The basic design will be a randomised controlled trial comparing FTT to and active dose matched control. We will also evaluate specific mediators and moderators of treatment effects by examining adolescent, family, and service provision variables that are likely to facilitate and hinder positive outcomes. Finally, we will examine costs and service provision data associated with the implementation of this intervention into a complex statutory setting responsible for managing youth offending in the UK.

Aims and hypotheses:
1. Outcomes: The main aim is to evaluate the effectiveness of FFT compared to a dose control condition that includes treatment as usual plus non-specific treatment for young people. It is hypothesised that FFT will be associated with greater improvements in family relationships, parent relationships to the offending youth, youth mental health, and finally, greater reduction in youth re-offending. Effects will be assessed over a one year period.
2. Outcomes across diverse adolescents and their families: The second aim is to assess the effects of the program across the various adolescent, family, and quantity and quality of service provision variables.
3. Mechanisms of change: The third aim is to assess what are the essential 'active' ingredients of the programmes in terms of changes in family, and parent and child behaviours in the domains that FFT explicitly targets for change. It is hypothesised that change in family process variables and the parent relationship with the target youth are necessary conditions for change in ongoing youth antisocial behaviour.
4. Success of service provision in the UK YOT system; suitability for dissemination: Finally, we will assess the total and relative social acceptability, reach, and cost benefits of FFT into the UK YOT system. What proportion of parents can be induced to attend, that is it suitable for a minority or the majority of families? What are the characteristics of those who don't take up the programme? Are their differences across programmes? What is the acceptability of the programmes to parents, frontline practitioners, senior managers and commissioners? How well can practitioners be trained to implement the programme and thus produce change?

Ethics approval

King’s College London Research Ethics Committee, 28/03/2008, ref: CREC/07/08-141

Study design

Two-arm randomised controlled trial

Primary study design

Interventional

Secondary study design

Randomised controlled trial

Trial setting

Other

Trial type

Treatment

Patient information sheet

Not available in web format, please use the contact details below to request a patient information sheet

Condition

Conduct disorder, antisocial behaviour

Intervention

1. Functional Family Therapy group:
FFT works by working improving outcomes for families by addressing a number of factors that may contribute to the likelihood of a young person offending. Key amongst these are family circumstances and styles of parenting, which have been shown to be correlated strongly with delinquency and anti-social behaviour.

FFT aims to assist young people and their families to make meaningful changes in their functioning. This is accomplished by reducing negativity in the family and focusing on significant yet obtainable behavioural changes that will have a lasting impact on family relationships. These changes may have the immediate effect of changing a specific functional problem or difficulty and the longer term effect of empowering a family to continue to apply changes to future circumstances. A key duty of the therapist is to assume some of the functional duties of a "family case manager" to assist the family to identify and interact with relevant community resources that will help the family sustain their changes over time.

In this way, FFT aims to increase the family's capacity to manage problems, enabling a reduced reliance on service providers and their resources. It is a home-based assertive outreach model. The intervention process involves, on average, 12 one hour sessions over a period of 3 to 4 months delivered in the family home, with more complex cases requiring up to 30 hours of direct service. Thus young people in the FFT group will receive treatment as usual as normally provided by agencies and FFT.

Experienced family therapists have been hired to provide the intervention. They have been trained to the highest standard by the originator of the programme, Professor James Alexander, and will receive weekly supervision. Given that the SAFE project is an effectiveness trial the therapists will be located within, and will work as part of, the Brighton and Hove Youth Offending Team.

2. Dose Control group:
The use of standard, no-treatment control groups in behavioural intervention research can be problematic for a number of reasons. In addition to ethical and legal considerations of denying treatment to this population, the use of no-treatment control groups may not control for a number of confounding factors. In particular, a treatment group may have improved outcomes, not because of the effectiveness of the intervention, but because they receive additional attention from a therapist. Whilst the use of dose control groups has become more widespread in clinical trials of interventions for depression it has been sorely lacking in evaluation trials of parenting programmes or family therapy.

This study will address these issues by including treatment as usual as normally offered by agencies, and incorporating additional service provision in the control group that matches the amount of time and attention received by participants receiving FFT. Thus adolescents and their families will receive treatment as usual and an additional 12 hours of additional contact with, and support from, a caseworker not trained in FFT.

Intervention type

Other

Phase

Not Applicable

Drug names

Primary outcome measures

Young people's offending and antisocial behaviour:
1. Young Person (YP) offending, reoffending and breach of orders: youth offending service databases and police national database (measured at T1, T3, T4)
2. YP delinquency, conduct disorder and antisocial behaviour:
2.1. Smith and McVie Self-report Delinquency questionnaire (measured at T1, T3, T4)
2.2. Adolescent Parent's Account of Child Symptoms interview (measured at T1, T3, T4)
2.3. Strength and Difficulties Questionnaire (Primary Caregiver [PC] and YP self-report) (measured at T1, T2, T3, T4)
2.4. Antisocial Process Screening Device (parent and self-report) (measured at T1, T3, T4)

Timepoints:
T1: baseline assessment directly before randomisation and treatment
T2: 3 months post-randomisation
T3: 6 months post-randomisation
T4: 18 months post-randomisation

Secondary outcome measures

1. Peer relationships:
1.1. Peer delinquency: Smith and McVie Peer Delinquency Self-Report Questionnaire (measured at T1, T3, T4)
1.2. YP peer relationship problems: Strength and Difficulties Questionnaire (PC and self-report) (measured at T1, T2, T3, T4)
1.3. Child prosocial behaviour: Strength and Difficulties Questionnaire (PC and self-report) (measured at T1, T2, T3, T4)
2. Young peoples mental health:
2.1. YP internalising disorders: Strength and Difficulties Questionnaire (PC and self-report) (measured at T1, T3, T4)
2.2. YP hyperactivity/inattention: Strength and Difficulties Questionnaire (PC and self-report), Conner's Rating Scale for ADHD (measured at T1, T4)
2.3. Child autistic spectrum disorder: the Childhood Asperger Syndrome Test (PC questionnaire) (measured at T1)
2.4. Child psychosis: Psychotic-Like Experiences Questionnaire (self-report) (measured at T1)
2.5. Child post-traumatic stress disorder: Impact of Events Scale (self-report questionnaire) (measured at T1)
3. Young peoples education attainment and intelligence quotient (IQ):
3.1. Child school attendance, exclusions, academic achievement and statements of special educational needs: school records (measured at T1, T3, T4)
3.2. IQ: Wechsler Abbreviated Scale of Intelligence (measured at T1)
4. Family relationships and parenting:
4.1. Child attachment: Child Attachment Interview (measured at T1, T4)
4.2. Parental relationship: the Index of Marital Satisfaction Questionnaire and Conflict Tactics Scales Questionnaire (PC questionnaire) (measured at T1, T3, T4)
4.3. Disorganisation and chaos in the home environment: Confusion, Hubbub and Order Scale (PC questionnaire) (measured at T1)
4.4. Parenting behaviour: Alabama Parenting Questionnaire (PC [measured at T1, T2, T3, T4] and other parental figures in family [measured at T1, T3, T4], child self-report [measured at T1, T3, T4])
4.5. Parental attribution of intent: when interviewing primary caregivers we ask them to imagine their child doing a number of things that could be regarded as positive or negative. We then ask them why they think their child is doing these things. This allows us to assess their attribution of intent: whether they think their child's behaviour is driven by positive or negative motives (PC interview) (measured at T1, T2, T3, T4)
4.6. PC self-efficacy: Carer Confidence Questionnaire: 9 item PC questionnaire assessing sense of control and agency (PC questionnaire) (measured at T1, T2, T3, T4)
4.7. Reflective functioning: Parent Development Interview (PC interview) (measured at T1, T3, T4)
4.8. Expressed emotion: Psychosocial Assessment of Childhood Experiences Coding Scheme (PC interview) (measured at T1, T3, T4)
4.9. PC and YP interaction: video observations coded with Problem Solving & Dyad Interaction Coding Scheme (video observation of YP and PC) (measured at T1, T3, T4)
5. Parental mental health:
5.1. PC depression: the Depression, Anxiety and Stress Scales (PC self-report questionnaire) (measured at T1, T3, T4)
5.2. Parental history of antisocial behaviour: Mother's and Father's Antisocial Personality Questionnaire (PC questionnaire) (measured at T1)
5.3. PC alcohol and drug misuse: The Alcohol and Drug Use Disorders Identification Tests questionnaire (PC self-report) (measured at T1)
6. DNA:
6.1. Child DNA: collected with cheek swab (measured at T1)
6.2. PC DNA: collected with cheek swab (if biological parent) (measured at T1)
7. Health economics:
7.1. Service use and demographics: Client Service Receipt Inventory (PC interview) (measured at T1, T3, T4)
7.2. Quantity and type of treatment as usual received by young people: Youth Offending Service and other agency databases (measured at T1, T2, T3)
8. Process variables:
8.1. FFT therapists' adherence to model and competence: assessments undertaken by FFT trainers (measured at T1, T2, T3)
8.2. Client-therapist alliance: standard FFT including questionnaires (measured at T1, T2, T3)

Timepoints:
T1: baseline assessment directly before randomisation and treatment
T2: 3 months post-randomisation
T3: 6 months post-randomisation
T4: 18 months post-randomisation

Overall trial start date

01/07/2008

Overall trial end date

01/12/2011

Reason abandoned

Eligibility

Participant inclusion criteria

1. Young people aged 10 - 17 years (either sex) and their families (mainly primary caregivers but also others with a parental role)
2. Clients of a CYPT agency being seen for offending or antisocial behaviour receiving a casework-based intervention

Participant type

Patient

Age group

Child

Gender

Both

Target number of participants

100 young people and their families

Participant exclusion criteria

1. Young people with a major developmental disability
2. Looked after young people (this does not include long term foster placements)
3. Young people not living in the family home

Recruitment start date

01/07/2008

Recruitment end date

01/12/2011

Locations

Countries of recruitment

United Kingdom

Trial participating centre

PO85 De Crespigny Park
London
SE5 8AF
United Kingdom

Sponsor information

Organisation

Kings College London (UK)

Sponsor details

c/o Richard Barnard
Institute Secretary
Institute of Psychiatry
De Crespigny Park
London
SE5 8AF
United Kingdom

Sponsor type

University/education

Website

http://www.kcl.ac.uk

Funders

Funder type

Government

Funder name

Department of Children, Schools and Families (UK) - The National Academy for Parenting Research

Alternative name(s)

Funding Body Type

Funding Body Subtype

Location

Results and Publications

Publication and dissemination plan

Not provided at time of registration

Intention to publish date

Participant level data

Not provided at time of registration

Results - basic reporting

Publication summary

Publication citations

Additional files

Editorial Notes