Time trends in childhood mental health problems: a 40-year cross-cohort comparison

Many people and professionals are worried that child mental health problems have become more common in recent years. For example, more children today are diagnosed as having Attention Deficit Hyperactivity Disorder (ADHD) than in the past. However, this could be due to better understanding of the condition by teachers and doctors. It is important to understand whether or not rates of ADHD, and of other child mental health problems such as emotional symptoms and behavioural problems, have actually increased in the population as a whole. This requires data from unselected representative national samples assessed in the same way over time. In our previous research we have shown that emotional and behavioural problems in adolescence have in fact become increasingly common in recent years. At present we don't know whether the same holds true for mental health problems in pre-adolescent children. The study will answer this question by using data on the mental health of seven-year-olds from five large UK cohorts studied between 1965 and 2008. Each study has used identical or closely comparable child mental health screens.

Another important issue is what might explain any changes in children's mental health. Very few studies have looked at this, because comparable information about possible contributory factors and mental health is only rarely available for the same individuals when combining cohorts across time. Mental health trends in adolescence are partly explained by increasing rates of parental mental health problems, by changes in youth drinking and drug use, and by increasingly stressful educational experiences. This type of research now needs to be undertaken in younger children. We will do this by testing whether rates of early known risk factors have changed across the five cohorts, and how such changes have contributed to trends in child mental health. We will focus on changes in pre- and peri-natal risks (e.g. prenatal maternal smoking, prematurity and low birth weight), in early neurodevelopmental risk factors (e.g. head injury and epilepsy), and early family adversity (e.g. parental involvement, parental divorce and family poverty).
Our third main aim is to test whether the impact of child mental health problems on children's later health and wellbeing has changed. In particular, we want to see whether any increases in mental health problems at age seven foreshadow increases in mental health problems in later childhood and adolescence. We also want to examine the consequences for children's psychosocial adaptation, including their peer relationships and educational success. This is made possible because three of the cohorts are being followed as they grow older. Information from middle childhood and adolescence is now available for two of these cohorts, previously assessed at age seven in 1965 and the late 1990s. Longitudinal outcome data will also soon become available for the most recent cohort of children who are currently being assessed at age eleven.

Understanding whether rates of child mental health problems have changed, the reasons behind trends in child mental health, and the consequences for children's later development has important practical consequences. This information in crucial for better planning of services, for influencing national and local policy and priorities, and for identifying targets for intervention to improve the mental health of future generations of children.

Accurate understanding of mental health trends requires the comparison of representative unselected population samples assessed in the same way at different points in time. Cross-cohort comparisons have shown that rates of adolescent emotional and behavioural problems have increased in recent decades and identified some possible contributory factors. Evidence on mental health trends in pre-adolescent children is lacking. The study objectives are i) to test trends in child emotional, conduct and ADHD problems over a forty year period (1965-2008), ii) to test trends in known risk factors (pre- and peri-natal, neurodevelopmental, and early family adversity) and their contribution to trends in child mental health, iii) to assess the impact of child mental health problems on later psychosocial functioning for earlier and later cohorts. The study combines data across five UK population cohorts of seven-year-olds - the National Child Development Study (NCDS, assessed in 1965, N = 13500), the Avon Longitudinal Study of Parents and Children (ALSPAC, assessed in 1998/9, N = 8500), the ONS British Child and Adolescent Mental Health Surveys (assessed in 1999, N = 1000 and in 2004, N = 650), and the Millenium Cohort Study (MCS, assessed in 2008, N = 13500). Comparable parent-rated symptom screens were used in all five studies - the Rutter A scale in 1965 and the Strength and Difficulties Questionnaire (SDQ) in the latter four cohorts. This study draws on a core set of comparable items across the two scales. Minor differences in response format and wording will be addressed using calibration data. The cohorts include comparable assessments of proposed risk factors. Impact of child mental health problems on functioning will be assessed using parent and teacher symptom scores at age 11 (NCDS, ALSPAC, MCS) and age 16 (NCDS, ALSPAC), and exam attainment at age 16 (NCDS, ALSPAC).

One in ten children and adolescents suffer from clinically significant emotional and behavioural disorders. Mental health problems impair children's quality of life and place a burden on families and schools. They also show strong continuities with mental health problems in adolescence and adulthood, and have consequences for young people's future life chances, health and education. It is therefore a key social, economic and health priority to develop policies and practical interventions that reduce the incidence and burden of child psychopathology.

It is of major concern that evidence now highlights that mental health problems have become substantially more common in recent decades, at least in adolescence. We do not yet know whether similar mental health trends are also observed in younger children. Direct tests of possible explanatory factors are rare. Some research has begun to identify possible contributors to mental health trends, but have mainly focused on risks of particular relevance to adolescents (e.g. substance use or exam stress). We do not know the implications of changes in pre-/peri-natal, neurodevelopmental, or early family risk factors for trends in child mental health. Consequences of changes in child mental health for functioning and future psychosocial adjustment also need to be better understood.

Our research addresses each of these important knowledge gaps, and findings will be of direct benefit to national and local policy makers, as well as more generally to health and social care professionals working with young children. Research has the potential to make significant impact on policy and practice. Knowing if mental health problems have increased for pre-adolescent children will inform whether policy efforts should already be focused on families with younger children. More generally, establishing changes in population prevalence contributes to determination of national and local government policy priorities. In this way, our previous research contributed to efforts to make youth mental health a national policy priority. Third, identification of plausible contributory factors underlying child mental health trends is a pre-requisite for evidence-based policy and prevention efforts to improve the mental health of future generations of children.

The applicants have a strong track record of engagement with policy makers, health and social care professionals, and others to maximise the impact of findings from the time trends research programme. Findings have also been communicated to the wider public via national and international media. The achievement of impact of this study will again be facilitated through rapid dissemination of key findings via web-based briefing papers, presentations to policy and professional audiences, and via local and national media. As noted in the application, this strategy has achieved significant policy impact, with citations by numerous governmental (e.g. parliamentary select committees on education and health, NICE, Institute of Medicine, WHO) and non-governmental bodies (e.g. Children's Society, Mind, NSPCC).