Attention deficit hyperactivity disorder (ADHD) is a common, extremely disabling disorder which has major adverse squalae in childhood and later life. Available evidence suggests that both genetic and environmental risk factors are important in the aetiology of ADHD; how these influences, and their interplay, impact the course (i.e., trajectories) of ADHD symptoms is poorly understood.

The aim of this proposal is to further explore individual trajectories of ADHD symptoms over time and genetic and family environmental predictors of those trajectories. In addition, we aim to explore the influence of symptom age of onset and psychiatric comorbidity on the course of ADHD.

ADHD symptoms trajectories, on average, show declines from middle childhood through early adolescence (Langberg et al., 2008), although environmental stressors (in this study, the transition to secondary school) can lead to temporary reversals of these declines. Previous research using the ALSPAC sample has identified 4 latent class trajectories of ADHD symptoms over time: low risk, intermediate, childhood limited, and persistently impaired (Pourcain et al., 2011). Other research has found evidence for three symptoms trajectory classes: low; increasing in childhood, then decreasing; and decreasing during childhood and then increasing later in development (Malone et al., 2010). Individual trajectories of ADHD symptoms have implications for later adjustment; in the Malone et al., study, participants with the 3rd trajectory (decreasing symptoms during childhood, then later increases) reported an earlier onset of illicit drug use than the stable low trajectory class.

Little to no research has examined either genetic or family environmental influences on ADHD trajectories over time. Increased knowledge about the aetiology of these symptom trajectories in epidemiological samples is crucial for prevention efforts and provides preliminary foci for future research in clinical samples.

1). First, we propose to explore individual trajectories of ADHD symptoms over time and associations among initial symptoms and longitudinal course. Predictors of these trajectories will then be examined, including the family environment, genetic risk, age of symptom onset, and psychiatric comorbidity, specifically symptoms of CD and ODD.

2). We also propose to look at the COMT val158met genotype which has been shown to be associated with CD in samples with comorbid ADHD (Langley et al., 2010) and a (previously generated) polygenic risk score, following work finding significant association between risk score and comorbid CD in those with ADHD (Hamshere et al., submitted).

3). Using this genetic information, we are interested in looking at parental transmission of risk alleles to their child; if a child is exposed to a dysfunctional family environment and the mother has the risk genotype but does not pass that genotype on to her child, the course of ADHD symptoms over time should be less pathological if shared genetic liability were important, but not if environmental mediation was at work.

4). We additionally plan to examine moderation of genetic risk on longitudinal trajectories of ADHD symptoms by the family environment. The expectation is that children who both have genetic risk for ADHD and are exposed to dysfunctional environments will evidence steady high and/or increasing trajectories of ADHD symptoms over time.

These analyses would help elucidate predictors of the course of ADHD. Furthermore, we aim to replicate our findings from ALSPAC in our clinical sample of (n=800) children with ADHD which should provide further insight.

The polygenic risk score for ADHD and the COMT vala58met genotype that we propose to study are strong candidates for association with ADHD trajectories and associated conduct problems and are also available in the ALSPAC data set, the two main reasons why they have been proposed here. We recognise that other variants in these and other genotypic regions may also be linked to ADHD symptoms and, should funds be available for future genotyping, would be interested in extending our analyses.

Because our proposed study is similar to current work by Barbara Maughan on CD trajectories, we have been in consultation with Dr. Maughan and will maintain a close collaboration during this project to both enhance the quality of the proposed research and to ensure the distinctness and complementarity of our studies. The proposed project with also be conducted with the collaboration of Prof. Gordon Harold and Dr. Stephan Colinshaw at the University of Cardiff.

The proposed study will utilize the polygenic risk score for ADHD currently under development by Joanna Martin (ALSPAC project # B1342; planned completion date for deriving the polygenic risk score is Nov. 1st, 2012). The polygenic risk score is calculated by using the results (below a chosen threshold) of a genome-wide association study (GWAS) of ADHD cases vs. population controls to create a polygenic score for each individual in the target dataset (i.e. in ALSPAC). Thus, we propose to examine the effect of common genetic variants implicated in ADHD, in the form of polygenic scores calculated in ALSPAC based on a discovery sample of children diagnosed with ADHD, on trajectories of ADHD during childhood and adolescence.

Current data required:

1) Parent & Teacher reports DAWBA symptoms and diagnoses of ADHD, CD and ODD at all available assessment points

2) SDQ - subscale summary scores at all available time points

3) Child- and parent-report of antisocial behavior

4) Child and parent report of parental monitoring, parenting, and the parent-child relationship at all available assessments

Items: 6 months: KB554-KB586; 18 months: KD270-KD305, KD366-KD422; 24 months: KE010-KE021; 30 months: KF300-KF326, KF358-KF394, KF430-KF431; 42 months: KJ250-KJ268, KJ298-KJ334, KJ370-KJ459; 77 months: KP6000-KP6050, KP6210-KP6230, KP6270-KP6282; 103 months: KT3030-KT3098; 115 months: KU280-KU329, CCF104, 111, 118, 125, 133, 141, 149, 157, 165; 140 months: KW4000-KW4050, KW9020-KW9088; 169 months: CCR500-CCR550; 16.5 years: TC1000-TC1180

5) Age (parents & child at assessments)

6) Gender

7) BWT

8) Derived variables of Partner Affection, and Partner Aggression at each available time point (e.g., F593-F595 & F596-F598 from 8 week Mother Questionnaire)

9) SES & Mothers education

10) Mother & Father psychopathology - ADHD, CD, Anx. & Depression

11) Mother & Child genotypes

References:

Langberg, J. M., Epstein, J. N., Altaye, M., Molina, B. S. G., Arnold, L. E., Vitiello, B. (2008). The transition to middle school is associated with changes in the developmental trajectory of ADHD symptomatology in young adolescents with ADHD. Journal of Clinical Child and Adolescent Psychology, 37, 651-663.

St. Pourcain, B., Mandy, W. P., Heron, J., Golding, J., Smith, G. D., & Skuse, D. H. (2011). Links between co-occurring social-communication and hyperactive-inattentive trait trajectories. Journal of the American Academy of Child & Adolescent Psychiatry, 50, 892-902.

Malone, P. S., Van Eck, K., Flory, K., & Lamis, D. A. (2010). A mixture model approach to linking ADHD to adolescent onset of illicit drug use. Developmental Psychology, 46, 1543-1555.

Langley, K., Heron, J., O'Donovan, M. C., Owen, M. J., & Thapar, A. (2010). Genotype link with extreme antisocial behavior. Archives of General Psychiatry, 67, 1317-1323.

Hamshere, M.L., Langley, K. et al., High loading of polygenic risk for ADHD in those with comorbid aggression. Submitted.