Trajectories of Conduct Problems from Ages 3 to 11

Recent independent reports to the Government have highlighted the importance of devising interventions to promote the healthy development of children. In particular, conduct problems in early childhood have been identified as an important determinant of life chances, with wide-ranging and long-lasting individual and societal impacts. Conduct problems contribute strongly to health inequalities and are a key mechanism for the inter-generational transmission of disadvantage. The lifetime costs of conduct problems are high and even modest improvements are likely to produce a return on investment in early intervention.

While there is a growing consensus on the need for early intervention, devising an effective intervention strategy is not a straightforward task. Previous research suggest that children exhibit different developmental pathways of conduct problems, depending on their exposure to risk and protective factors which, in turn, has implications for longer term outcomes. Yet, few studies have examined trajectories of conduct problems from an early age using a contemporary longitudinal data set with a nationally representative sample of children.

We aim to fill this knowledge gap by investigating children's conduct problems using the Millennium Cohort Study (MCS). Our proposed project has five major objectives: (1) to investigate the prevalence of behavioural problems at ages 3, 5, 7 and 11 and their co-morbidity with other mental health conditions such as emotional problems and hyperactivity; (2) to assess developmental pathways over time and examine different patterns in the trajectories of conduct problems according to age of onset, persistence and severity; (3) to identify the relative importance of risk and protective factors at different ages; (4) to examine the cascading effects of conduct problems on other life domains; and (5) to inform policy makers and practitioners regarding the design and implementation of effective early interventions.

MCS covers a representative sample of around 19,000 children born in the UK in 2000/01 and has collected a large body of detailed information on the cohort children, their parents and their older siblings. We will focus on conduct problems from mild to severe detected by the Strengths and Difficulties Questionnaire (SDQ) (Goodman, 1997). There are several unique features of the MCS which we will use to create new insights. First, SDQ scores are available at ages 3, 5, 7 and 11 so we can examine variations in developmental pathways of conduct problems across time. Second, using parent and teacher-reported responses to the SDQ, we can compare responses by different reporters. Third, there have been few, if any, studies documenting ethnic differences in the development of conduct problems. MCS offers a unique opportunity to examine these differences, as it has been oversampled for ethnic minorities (about 16% of the sample). Fourth, information regarding SDQ scores of older siblings enables us to control for otherwise unobserved features of a family. Lastly, there is information on a wide diversity of risk and protective factors available in MCS, allowing us to explore cumulative and/or interactive effects and examining developmental change.

Improved understanding of trajectories in conduct problems is important not only from a research perspective but also because of its relevance for clinical practice, service provision and national mental health policy, and wider social policy. Through a strategic academic/third sector partnership between the IoE's Department of Quantitative Social Science and the Centre for Mental Health, we aim to disseminate findings in different contexts and to communicate effectively, well beyond the research community.

The main objective of the research is to improve understanding of the onset and development of serious conduct problems in childhood, with the ultimate aim of helping to reduce the occurrence of these problems through more effective early intervention. The range of potential beneficiaries from this work is very wide, reflecting the many different domains of life that are affected if these problems are allowed to persist, and the multiple risk and protective factors that may influence the development of conduct problems and their development over time.

For example, young children with severe conduct problems are very likely to engage in criminal activity in later life. Prevention of these problems thus offers clear benefits, both to society generally and to the criminal justice system, from future crime reduction. Less obviously, criminal justice agencies may also have a direct role in delivering early intervention as well as benefiting from it, e.g. by providing parent training programmes to prisoners and their partners, a group whose children are known to be at high risk of conduct problems. Early intervention thus goes well beyond the ambit of children's mental health services and may also require input from social services, schools, the criminal justice system, housing departments, maternity services, drug and alcohol agencies, and so on (Robins & Rutter, 1990) . The concentration of children with conduct problems in low-income families also implies a potential role for national policies of taxation and social security.

The general principle of early intervention is widely accepted, but devising and implementing effective strategies is a major challenge. Our research is relevant in a number of respects. About half of young children with conduct problems outgrow these difficulties before adolescence, but relatively little is known about the characteristics of those in whom problems persist. Better identification of this group would strengthen the targeting of interventions and their cost-effectiveness. Analysis of the risk factors causing persistence and of the protective factors promoting resilience will inform the appropriate selection and design of interventions. Analysis of trajectories including differences in age of onset will be relevant to the optimal timing of interventions and to whether interventions should be repeated at different points in childhood. To maximise the relevance, we will seek to identify and highlight those risk and protective factors that are not only potent in terms of their effect size but also potentially modifiable in a cost-effective way via policy and service provision.

Our project will also benefit the academic community, by examining the dynamics of childhood conduct problems using a contemporary longitudinal data set with a large representative sample of children. We will focus specifically on several areas of research which fill gaps in existing knowledge including: the gender and ethnic-specific pathways of conduct problems; the cumulative and interactive nature of risk and protective factors at different ages; and the development of a cascade model of how early conduct problems may develop into poor later outcomes. Our research will provide a greater understanding of the development, course and prognosis of conduct problems, relevant to a wide range of researchers working on children's mental health.

Finally, we hope that our findings will ultimately be of benefit to the many families with children whose life chances are compromised by conduct problems. About 5% of children suffer from conduct disorder at any one time, with a further 15% with sub-threshold problems. Altogether this implies 120,000 children at risk in each one-year cohort of children, or 0.75 million in the country's primary schools. Our research aims to strengthen the evidence base for policies and services which influence the mental health and well-being of these children and their families.