There is considerable evidence suggesting that a wide range of childhood and adolescent externalizing behaviors are prospectively associated with the subsequent development of alcohol use (AU) and alcohol use problems (AUPs), including several dimensions on which data have been collected within the ALSPAC study: e.g., conduct problems, temperament, personality, sensation seeking, and impulsivity. Clearly, we would not expect these various aspects of externalizing behavior to act independently, but rather to operate on AU and AUPs through a dynamic system of influences, both within and across time. For example, early temperament has been shown to influence later conduct problems, for which there are also known associations with aspects of personality, impulsivity, and sensation seeking. There is little clarity, however, regarding how mechanistic associations between these facets ultimately influence AU and AUPs in adolescence and adulthood. Prior to examining the structural nature of these relationships, we must first establish the factor structures within each of the dimensions, independently. So, for example, ALSPAC has collected information on conduct problems assessed across several points in time, and with multiple instruments (e.g., the Revised Rutter Parent Scale for Preschool Children at 42 months and repeated assessments of the Strengths and Difficulties Questionnaire at ages 6.5, 9.5, and 11.5 years). In order to assess development within conduct problems, we must first confirm that the two scales tap into the same underlying construct. Toward this end, one strategy that we propose is to identify a set of "anchor" items/behaviors that are assessed within both measures (e.g., is inconsiderate of others, steals, destroys/damages property, fights, bullies, lies/cheats). We can then subject this common set of variables to within-wave confirmatory factor analysis, allowing us to differentiate between items that yield consistent parameter estimates across time (equivalent factor loadings across assessments) and those for which associations with the underlying construct vary across time (i.e., some items may reflect behavior that is more or less indicative of conduct problems at different stages of development). Time-specific factor score estimates (FSEs) can then be derived by specifying a model which equates factor loadings across time for those items displaying factorial invariance, while freely estimating loadings at each wave of assessment for those items displaying developmental inconsistency. Moreover, we propose to implement these procedures across the different dimensions of externalizing behavior that have been conducted in ALSPAC. The resulting FSEs, which capitalize on consistency while also recognizing the potential for specific behaviors to increase/decrease in relevance across development, can then be used as the unit of measurement in models addressing the complex structural relationships between individual dimensions of behavior, general domains of risk, and alcohol-related outcomes. Accordingly, the broad goal of this set of analyses will be to characterize the measurement of, and inter-relationships between, different domains of externalizing behavior that have been assessed in ALSPAC, in order to use this information to study how different facets of externalizing behavior relate to alcohol use. We have previously had a conference call with Barbara Maughan (and Glyn Lewis) and discussed her interesting work on trajectories of conduct problems and externalizing behavior in the ALSPAC sample, so we would plan to work with her (and other interested ALSPAC collaborators) on these efforts, so as to ensure there is no duplication of effort across the projects. Since our primary interest is in understanding how externalizing behavior relates to alcohol use (with the latter being our primary outcome of interest), and her interest is in the nature of externalizing behavior itself (her primary outcome of interest), we believe that our projects will be synergistic rather than overlapping in scope.

We propose a program project that will permit the processing and analysis of all consented ALSPAC placentas that have follow-up data on at least one of the three outcomes proposed for the program project, as described below.

The program project will include an Administrative Core (details to be provided by ALSPAC).

A "Placental Core" will be headed by Dr Craig Platt, Senior... at BRI, with Dr Carolyn Salafia as Co-PI at Placental Analytics LLC. Dr Platt will be responsible for supervision and quality control of placental examination, photography, tissue sampling, and sample processing to wax blocks, Placental Analytics LLC will be responsible for image analysis of placental photographs, and preparation of all histologic slides deemed essential to study goals including routine (hematoxylin and eosin) and special (e.g., immunohistochemistry) stains, and will digitize all such slides in a standard and accessible format. At the completion of the program project, all data extracted from photographs of placentas and digitized slide files, plus the digitized slide files, slides and remainders of wax tissue blocks will be returned to ALSPAC/BRI (?) as a permanent archive.

Program subprojects:

Neurocognitive: "Placental structure and pathology and neurocognitive outcome"

PI Prof Jean Golding, Co-Investigators D Misra, J Miles, C Platt, C Steer, C Salafia

We propose to test the hypothesis that detailed markers within the placenta are related to specific cognitive, behavioural and psychological outcomes from early childhood to school leaving (see Annex, below). A range of statistical approaches including structural equation modelling, a method well known to the psychometric literature, will be used to determine the predictive value of the placental measures in regard to the outcomes. Particular attention will be paid to ways in which the placenta varies in regard to factors that have been shown to be related to adverse neurocognitive outcomes such as influenza in mid-pregnancy, alcohol in early pregnancy, maternal anxiety, maternal prenatal levels of lead and mercury, and paternal omega-3 intake. We will also assess the ways in which placental variation is observed in placentas of children with minor congenital defects (minor congenital malformations are found in excess among children with many different neurocognitive outcomes).

All the neurocognitive outcomes will be available, but the minor congenital malformations will need to be identified from the neonatal clinical records - for which this grant would budget to be extracted from existing medical records.

The placenta provides two routes of analysis that we propose will account for variation in neurocognitive development/developmental trajectory:

1. Placental branching growth is determined by the branching growth of its vascular tree. The "neurovascular hypothesis of disease" summarizes recent recognition that the gene families, growth factors and molecular signals that promote and maintain neuronal health are the same factors that promote and maintain vascular health. We propose that measurement of the placental vascular tree, in terms of 2D chorionic surface, 3D placental volume, chorionic surface vascular networks and finer architecture of the placental fetal and terminal villi correlate with fetal neuronal growth and health and will therefore account for variability in neurocognitive outcome/developmental trajectory.
2. Placental pathologies of inflammation and injury patterns that may indicate oxidative stress can be identified in clinically abnormal and in clinically "normal" pregnancies. These pathologies are measured by histologic features identified in tissue sections of the placenta. We propose that measurement of these features will provide quantitative estimates of fetal exposures to inflammation and oxidative stress, via fetal effects on the developing brain, will account for variation in neurocognitive outcome/ developmental trajectory.

Pulmonary: "Placental structure and pathology and pulmonary outcome"

PI Dr John Henderson, Co-Investigators D Misra, J Miles, C Platt, C Salafia

There is evidence from cohort studies that have measured lung function sequentially from childhood that decrements of lung volumes (FEV1) and flows (FEF25-75 and maximal flows at 75%, 50% and 25% remaining FVC) may become established by early childhood and then 'track' into adult life. The effect of this would be to diminish peak measures of lung function prior to the natural decline in lung function with age, thus hitting a threshold at which disease becomes manifest at an earlier age. This paradigm is compatible with additional insults, such as smoking, leading to more rapid decline of lung function & exacerbating early life deficits. Thus, there may be additive effects of early life factors and later insults leading to diseases such as COPD.

Birth cohort studies that have measured lung function shortly after birth have demonstrated differences in measures of airflow associated with different patterns of symptom presentation in early childhood (e.g. transient early wheezing is associated with decrements in V'maxFRC shortly after birth (before symptoms are manifest) but regresses to the mean during the first years after birth whereas children who develop wheeze in later childhood have normal V'maxFRC in infancy but lower values of FEV1 in later childhood than never-wheezers) and with in-utero exposure to maternal tobacco smoke - children exposed in utero have lower values of lung function after birth and are more likely to develop wheeze and asthma during childhood.

Following the work of Barker and others, who demonstrated that adult respiratory diseases and lung function (FEV1) is associated with birth weight, there is considerable interest in the notion that fetal growth restraint is associated with abnormalities of airway development during a critical window of exposure leading to canalisation of effects that persist through the lifecourse (but may manifest only in the context of a second insult, e.g. smoking or following the decline in lung function measures with ageing).

Again, the placenta provides two routes of analysis that we propose will account for variation in pulmonary function:

1. Placental trophoblast epithelial branching growth, determined by the branching growth of its vascular tree, parallels the branching of the pulmonary epithelium that is coordinated with the development of its parallel vascular tree. Major airways are laid down in the embryo/fetus at similar times as the major chorionic surface vessels, and more distal villous branching parallels later lung growth in utero. The close linkage of these two organ's development is highlighted by the difficulty in studying mutations of genes such as FGF and Notch that are key to lung development, as many of such muiutations are embryonic lethals because of failure of the development of the placenta. We propose that measurement of the placental vascular tree, in terms of 2D chorionic surface, 3D placental volume, chorionic surface vascular networks and finer architecture of the placental fetal and terminal villi correlate with fetal pulmonary growth and health and will therefore account for variability in pulmonary function.

2. Placental pathologies of inflammation and injury patterns that may indicate oxidative stress can be identified in clinically abnormal and in clinically "normal" pregnancies. These pathologies are measured by histologic features identified in tissue sections of the placenta. Both inflammation, especially when cytokines are inhaled in the course of intraamniotic infections, and oxidative stress may alter pulmonary branching morphogenesis. We propose that measurement of these features will provide quantitative estimates of fetal exposures to inflammation and oxidative stress that will account for variability in pulmonary function.

More specifically, ALSPAC has detailed respiratory history at approximately 12 monthly intervals from birth and has measured lung function by spirometry at 8 and 15 years, allowing associations of both attained lung function pre- and (mostly) post-puberty and trajectory of lung function growth during adolescence (with a wealth of background information on exposures during this period, such as passive and active tobacco smoke exposure, validated with measurements of tobacco metabolites (cotinine).

Subproject 3 "Maternal overnutrition and offspring fat mass, metabolic and vascular function"

Principal Investigator: Dawn Misra, PhD, Co-Investigators: D Lawlor, J Miles, C Platt, C Salafia.

Dr. Lawlor's current grant aims refer to "offspring" generally; we have explicitly separated out the neonatal (birth outcomes) within our aims given the focus on the placenta.

SPECIFIC AIMS

Specific Aim 1: To investigate the influence of maternal BMI, weight gain, and diet during pregnancy on placental anthropometric measures. Specifically, we will assess and study the following placental parameters: placental weight, thickness, diameters; umbilical cord length; shape; symmetry measures; chorionic vascular branching pattern). The following null hypotheses will be tested:

1A. Maternal BMI does not explain variation in placental size and shape.

1B. Maternal weight gain does not explain variation in placental size and shape.

1C. Maternal diet does not explain variation in placental size and shape.

Specific Aim 2: To determine whether and how placental size and shape influences birth size. The following null hypotheses will be tested:

2A. Placental size and shape are not associated with birth size (e.g. weight, birth weight ratio, length, ponderal index).

2B. Associations between maternal factors and birth size are mediated by variation in placental size and shape.

Specific Aim 3: To determine whether and how placental size and shape influence offspring growth. Offspring growth was measured at multiple time points beginning at age 1 year up to 15 years of age and includes measures of adiposity (DXA assessed fat mass and fat distribution), vascular function (blood pressure, pulse pressure and endothelial function), and metabolic function (fasting glucose, insulin and lipids). The following null hypotheses will be tested:

3A. Placental size and shape are not associated with offspring childhood adiposity.

3B. Placental size and shape are not associated with offspring childhood vascular function.

3C. Placental size and shape are not associated with offspring childhood metabolic function.

3D. Associations between placental size and shape with offspring outcomes are mediated by effects of the placenta on birth size.

Specific Aim 4: To investigate the role of genetic variation on placental size and shape. DNA was extracted from mothers and offspring in the parent study. The following null hypotheses will be tested:

4A. Genetic variation does not explain variation in placental size and shape.

4B. Interactions between genes and the selected maternal factors (BMI, weight gain, diet) do not explain additional variations in placental size or shape.

Specific Aim 5: To develop and validate a placental index to identify offspring at risk for adverse outcomes with regard to adiposity, vascular function, and metabolic function.

Pilot analyses for the program subprojects will be generated using the placental data already collected from the Children in Focus subset (CIF, N=1050), and intermediate neurocognitive outcomes (TBD by Prof Golding), intermediate pulmonary function outcomes (TBD by Dr Henderson), and BMI, waist & fat & lean mass (DXA) at age 9 as previously determined by Dr Lawlor. These pilot analyses will provide strong support for our (generally well received) October submission, and we are optimistic about receiving funding based on our July resubmission. Dawn Misra and Jeremy Miles will work together on data analysis for the 3 subprojects, with Colin Steer additionally on the neurocognitive subproject.

Annex

1. Temperament and behaviour of child: Infant temperament using the Carey

at ages 6 and 24 months; EAS temperament scale at 42 months;

Behaviour measures of hyperactivity, antisocial/conduct problems, peer

problems, emotional behaviour and prosocial behaviour as well as overall

behaviour difficulties, measured at ages 4 - 13 years

2. Child development: mental and motor development measured annually

between 6 months and 4 years

3. Cognitive measures: IQ using the WISC at age 8; short term memory;

executive function;

4. Speech and language: Vocabulary measured from 15 months to 3 years;

Grammatical understanding; Phonological problems; Use of speech

tharapist

5. Motor coordination: Balance tests; fine motor coordination; Gross motor

coordination; Functional motor impairment, including DCD

6. Childhood growth: Height; Weight; Head circumference; Hip

circumference; Waist circumference; Body mass index - all at various time

points from 4 months to 15 years

7. Educational attainments: Results of SATS tests at 7, 11 and 14; GCSE

results; Results of ALSPAC's own tests ofReading, Spelling, Mathematics,

and Science understanding.

8. Attention and hyperactivity: Hyperactive and inattentive behaviour using

questions to parents and teachers from 4 to 11 years and to the children

at 13.

9. Hearing ability: Tests of hearing under standardised conditions at ages 7,

9 and 11 - using audiometry and otoacoustic emissions; Hearing of

speech discrimination at 3.5 and 5 years

10. Visual ability: Tests of visual acuity, stereoacuity, and other features of

visual ability at various ages from 8 months to 11 years

11. Other sensory outcomes (taste and smell): Taste test (PROP) results at

age 10; Smell test results at age 11

12. Childhood psychiatric disorders: Autism; ADHD; ADD; Anorexia nervosa;

Bulimia; Depression; Anxiety disorders; Conduct disorder;

We propose to test the hypothesis that genetic variation at two loci will reduce the likelihood of stopping smoking during the perinatal period. We expect these effects to be independent and will not investigate them in combination.

The first SNP (rs1051730) is located within the CHRNA3/A5 gene cluster, which encodes the nicotinic receptor alpha-3 and alpha-5 sub-units. Itwas originally reported to be associated with nicotine dependence in 2008 (Thorgeirsson et al., 2008), and this association has subsequently replicated in most (but not all) studies published since. It has also been investigated in relation to smoking cessation (e.g., Freathy et al., 2009). The National Institute of Drug Abuse (NIDA) has formed a working group to investigate the strength of evidence for the association of SNPs in chromosome 15 (including rs1051730) with smoking cessation. We propose to share the ALSPAC data published by Freathy and colleagues with this collaborative initiative, given the importance of including large, high-quality data sets of this kind.

The second SNP (rs4680) is located within the COMT gene, which encodes the catechol-O-methyltransferase enzyme, responsible for the degradation of extracellular dopamine, in particular in the prefrontal cortex. It has been widely investigated across a range of psychiatric phenotypes. We have recently reported an association between this SNP and smoking cessation in two clinical trials (Munafo et al., 2008; Johnstone et al., 2007), but subsequent attempts at replication have been inconsistent. One study found evidence of association in two independent samples of females only (Collila et al., 2005). Given evidence that oestrogen modulates COMT gene transcription (e.g., Jiang et al., 2003), it is possible that any effects of COMT genotype may be stronger in females.

Additional data on maternal depression over the same period are requested in order to explore whether any genetic associations with smoking cessation also predict change in depression symptom score, given the uncertainty regarding the direction of causation between cigarette smoking and depression.

Specific hypotheses:

1. The rs1051730 T allele will be associated with a reduced likelihood of smoking cessation;

2. The rs4680 G allele will be associated with a reduced likelihood of smoking cessation.

ALSPAC phenotypes required:

1. Maternal smoking status before, during and after pregnancy;

2. Maternal daily cigarette consumption and years smoked;

3. Maternal depression symptoms (EPDS) before, during and after pregnancy;

4. Maternal ancestry, age, educational attainment.

ALSPAC genotypes required:

1. rs1051730;

2. rs4680.

Aims

The main aim of the proposed study is to investigate the role of prenatal alcohol exposure on offspring cognition, neurological and motor development. I will also study maternal drinking patterns and how they influence offspring's alcohol behaviour including the onset of problem drinking, both directly and indirectly through a chain of biological effects on development and their consequences on educational attainment and choice of peer groups. The focus will be on causal effects and intergenerational determinants of health and health-related behaviours.

BACKGROUND

Depression is much commoner in women than men (Weissman, 1995). The high prevalence of depression in women is most likely due to a combination of gender-related differences in cognitive styles, certain biologic factors and a higher incidence of psychosocial and economic stresses in women (Korstein,1997). Depression during the perinatal period is common and it is debated if it is of a different type. Perinatal depression is often divided into antenatal and postnatal depression.

Postnatal Depression: Depression up to one year after delivery is called postpartum depression or postnatal depression. Postnatal depression is a mental and behavioural disorder associated with the puerperium, typically commencing within 6 weeks of delivery according to the International Classification of Disease-ICD-10(World Health Organisation, 1992). It is characterised by episodes of irritability, guilt, exhaustion, anxiety, sleep disorders and somatic symptoms which have a disabling effect on mothers, children and families of sufferers, following childbirth (Green et al., 1990). The cause of postpartum depression remains unclear with diverse researchers suggesting a multi-factorial aetiology (Ross et al., 2004). However, epidemiological studies strongly support the importance of psychosocial variables (O'Hara & Swain 1996, Beck 2001). In particular, a recent meta-analysis of 84 studies showed several risk factors including low levels of social support, stressful life events, childcare stress, low self esteem, low income, and marital dissatisfaction (Beck 2001). Buuljen (2007) concluded that one of the most notable explanations for PND was the lack of social support.

Social support: Social support is defined as help in difficult life situations. Social support is a concept that is generally understood in an intuitive sense, as the help from other people in a difficult life situations. There are different types and sources of social support. House described four main categories of social support: emotional, appraisal, informational and instrumental (House, 1981). Based on Barrera's (1981,2000) theoretical model, social support in the perinatal period can be defined as activities directed at assisting pregnant women and new mothers in mastering emotional distress, sharing tasks, giving advice, and teaching parenting skills, including the provision of material aid.

Social support seems to be one of the most important protective factors against depression in women at all periods of their lives; lack of social support has been consistently associated with depression in women of childbearing age. In order to understand better the relationship between social support and postnatal depression in women, I did a systematic review.

Social support and postnatal depression: In my sytematic review I was able to identify 20 papers that used a longitudinal design to assess the association between social support in pregnancy and postnatal depression. Among those that measured social support during the antenatal period, Brugha et al (1998), Glasser et al (2000), Bernazzani at al (1997) and Webster at al (2000) measured social support at less than 28 weeks of gestation; whereas Rahman et al (2003) and Chee et al (2008) took their measurements at the third trimester, between thirty four to forty weeks of gestations. Brugha et al (1998), Glasser et al (2000) and Webster et al (2000) found strong association between low social support at antenatal period and high EPDS score at postnatal period. Rahman et al (2003) found that support related to childcare by at least one of family member was associated with less PND, and the finding was almost similar to the study done by Chee et al (2000) when she found low instrumental support was associated with less PND but not for emotional support. Therefore from the studies we noticed that the measurement of social support was quite general when it was measured at early gestations, but when it was measured at late gestations the authors started to focus on type of social support.

Therefore my proposed study will be base on the objectives below.

OBJECTIVES

1. To study the association between social support in pregnancy and postnatal depression

2. To investigate the association between different types of social support in pregnancy and postnatal depression

3. To study if other psychological and social factors in pregnancy may contribute to postnatal depression in the presence of good social support.

METHODOLOGY

The proposed project will benefit from the rich and extensive longitudinal data obtained by Avon Longitudinal Study of Parents and Children (ALSPAC). ALSPAC is an ongoing population based study. This study will investigate the importance of social support in the ALSPAC mothers during pregnancy and the postnatal period.

Measures:

1. Depressive symptoms (mothers); antenatal and postnatal

2. Social support and social network measures

3. Personality

4. Religiosity

5. Locus of control

6. Life events

Analysis plan:

We will use complete datasets for the variables of interest. Initially I will perform a descriptive analysis of the main variables of interest during pregnancy and one year of postnatal period. I will then perform some basic cross-tabulations between social support measures during pregnancy and postnatal depression. Chi-square tests for these associations will be estimated. I will then examine the possible influence of confounding variables such as personality and life events. After this descriptive phase I will perform multivariate regression analysis to examine the association between postnatal depression and social support in pregnancy before and after adjusting for possible confounders. Potential moderating and mediating factors will also be examined. Finally I will explore applying other statistical approaches using more sophisticated models such as structural equation modelling.

Purpose:

To determine the extent that sedentary behaviour and measures of obesity track from childhood to adolescence.

Background:

Tracking of sedentary behaviour from childhood to adolescence has not been extensively studied, and few studies have tracked accelerometry derived sedentary behaviour (1). Using objectively acquired sedentary time at repeated time points is important since the reliability of self-reported sedentary behaviour is likely to vary with age. The study by Baggett et al. (1) found fair-to-moderate tracking of sedentary behaviour, from 12 to 14-years, in sample of girls (kappa=0.17). A second study regarding accelerometer derived sedentary time found that, from 6 to 9-years, there was moderate tracking (r=0.37-0.52) (2).

The use of DXA derived fat mass as a measure of obesity is highly advantageous over certain anthropometric measures of obesity, as adolescence is a period when lean mass is increasing in addition to fat mass (especially among males). The findings by Cheng et al. (3) found that the tracking of fat mass, from 11 to 18-years, was good (r=0.65); however, only Finnish girls (n=396) were included in the analysis.

Dependent Variable:

- Sedentary Behaviour (Tertiles)

- Obesity [Percentiles: 90th (obese), 80th (overweight) & 10th - 79th (normal)]

Indpendent Variables:

- Age (time)

Covariates:

- Maturation (Tanner)

- Parental occupation (SES)

- MVPA

Data Analysis:

- General Estimating Equations (GEE)

- Analyses will be conducted by gender (due to the different stages of maturation) and by SES (to determine the extent of tracking according to SES)

Target Journals:

MSSE

British Journal of Sports Medicine.

1. SUMMARY

Conduct problems (CP) include behaviors such as bullying/intimidating, initiating physical violence towards others, assault (stealing with confrontation), theft, vandalism, and running away from home. An estimated 5-10% of youth engaging in CP follow an early onset and persistent (EOP) trajectory, and account for over 50% of crime within a given community (Moffitt, 2006). The cost to both victims and communities is considerable (Foster). Nevertheless, well designed longitudinal studies converge on the finding that the development of CP is a heterogeneous process: at least two other trajectories have been consistently identified (Barker & Maughan, 2009; Odgers et al., 2007; 2008; Raine et al., 2005), neither of which appear to be completely free of risk in childhood nor adjustment problems thereafter. Preliminary findings from epidemiology studies suggest that youth may encounter environmental risk at different developmental periods, and that these risks (or protective factors) may interact within an individual's genotype to confer the onset or offset of CP.

Using previously published CP trajectories from ALSPAC (Barker & Maughan, 2009) as an outcome, we propose to examine environmental risk exposures, from gestation to the teens, and potential gene-to-CP effects, and gene-environment interactions. With regard to candidate polymorphisms, we propose to examine certain single nucleotide polymorphisms (SNPs) and variable number tandem repeats (VNTRs) that have been shown to associate to variation in stress response and/or CP. We also propose to examine a meta-analytic genome wide association study (GWAS), between ALSPAC and the Twins Early Development Study.

2. BACKGROUND

Well designed longitudinal studies converge on the finding that the development of CP is a heterogeneous process: beyond the EOP subtype, at least two other trajectories of CP have been consistently identified (Barker & Maughan, 2009; Odgers et al., 2007; 2008; Raine et a., 2005), a childhood limited (CL) and an adolescent-onset (AO) subtype. Taken together, the results of these studies suggest heterogeneity in development poses a serious challenge for clinicians, interventionists, and researchers alike. Understanding etiologic pathways underlying these distinct trajectories is a significant public health concern and efforts should be focused not only on prevention, but on ensuring that children who present CP have the best possible outcomes.

Gaps in the current knowledge on the etiologies of these CP pathways are at least three. First, to date, studies typically have examined sets of risk at specific (or aggregated) developmental periods, but rarely have these risks incorporated the prenatal period and early postnatal periods, crucial times in child development that may bear on the timing and impact of important risks on the development of biological systems, and hence an individual's susceptibility to subsequent environmental risk exposure(s). Second, no matter the severity and duration of risks experienced by children, evidence highlights individual differences in response, with some individuals apparently protected, or more vulnerable, to later adverse outcomes (Rutter, 2009). That is, there is consensus that risk factors are likely to be complex and involve interplay amongst genetic vulnerabilities and environmental risk exposure. The third limitation is that most studies to date have examined the EOP pathway, highlighting the need for further studies to examine etiologies underlying the CL and AO pathways. Indeed, differences in the development of CP amongst these trajectories suggests that the variation and timing of environmental risk, and presumably genetic vulnerabilities (or protective factors), as key features for refining current knowledge of these taxons.

We propose to test the extent to which variations in development of CP is related to the developmental interplay of genetic and environmental child-based risks. Specifically, we propose to conduct theoretically driven analysis where the research focus includes environmental risk exposure during pregnancy (18 wks and 32 weeks), early postnatal (birth to age 2; age 2 to age 4), school entry (age 7-8), adolescence (age 15-16), as well as candidate genes.

3. METHODS

3.1 Overview

Phase 1. Examine developmental continuity (and discontinuity) of environmental risk exposures during the following developmental periods: antenatal, early childhood, primary school and adolescence.

Phase 2. Examine both direct effects (polymorphism-to-CP trajectory) and interactions between polymorphisms and the environmental risks (i.e., examine if the relationship between a polymorphism and a trajectory is moderated by environmental risk exposure). SNPS and VNTRs prioritized via previous CP studies and studies on variation in stress response will be typed in the Avon Longitudinal Study of Parents and Children (ALSPAC, PI: G. Davey-Smith; Golding, Pembrey, & Jones, 2001). These data will be used to test hypotheses of direct gene, gene-environment interaction and haplotype-environment interactions where the outcome is the the odds of following the different CP trajectories. The availability of prospective information on environmental exposure in this sample will allow hypotheses to be tested about the impact of timing of environmental stress for these different CP trajectories.

Phase 3. GWAS meta-analysis. I have negotiated access to the Twins Early Development Study (TEDS, PI R. Plomin; Trouton, Spinath, & Plomin, 2002). Dr. Beate Glaser, of ALSPAC, has agreed to collaborate and wil estimate the GWAS on ALSPAC. Dr. Glaser will also combine GWAS estimates obtained from TEDS with those from ALSPAC.

3.2 Data collection.

We propose to genotype SNPs and VNTRs, as well as haplotypes of CRH-R1 (e.g., Lake et al., 2003) in ALSPAC.

3.3 Existing Data Required

Environmental risks are located in the previous section. The SNPs and VNTRs we propose to genotype are located in the next section.

3.4 Data Analysis.

Phase 1: We will estimate and examine the configuration of risk within each development period by way of Latent Class Model (McCutcheon, 1987). Continuity of environmental risks will be examined via latent transition models (Muthen & Muthen, 2001-2006).

Phase 2: Examination of gene-environment processes

Phase 3: GWAS meta-analysis

3.5 Work already completed

We have already published a paper on the trajectories and antenatal and early postnatal risks associated to the EOP vs CL pathways (Barker & Maughan, 2009). We will soon submit a follow-up to this paper (Barker, Bonamy & Maughan, in preparation) for approval to the ALSPAC exec. These two papers create the substantive background for this application.

References

Barker, E. D., & Maughan (2009). Differentiating early-onset persistent versus childhood-limited conduct problem youth. American Journal of Psychiatry, 166, 900-908.

Barker, E. D., Bonamy, B., & Maughan, B. (in preparation). Time-varying covariates of conduct problem trajectories.

Golding, J., Pembrey, M., & Jones, R. (2001). ALSPAC--the Avon Longitudinal Study of Parents and Children. I. Study methodology. Pediatric and Perinatal Epidemiology, 15, 74-87.

Lake, S. L., Lyon, H., Tantisira, K., Silverman, K. E., Weiss, S. T., Laird, N.M., Schaid, D.J. (2003). Estimation and Tests of Haplotype-Environment Interaction when Linkage Phase Is Ambiguous. Human Heredity, 55, 56-65.

Moffitt, T. E. (2006). Life-course persistent versus adolescence-limited antisocial behavior. In D. Cicchetti & D. J. Cohen (Eds.), Developmental Psychopathology (2nd ed., Vol. 3: Risk, disorder, and adaptation, pp. 570-598). NY: Wiley.

Muthen, L. K., & Muthen, B. O. (1998-2006). Mplus. Statistical analyses with latent variables. User's guide (4.1 ed.). Los Angeles: Muthen & Muthen.

Odgers, C. L., Caspi, A., Broadbent, J. M., Dickson, N., Hancox, R., Harringthon, H., et al. (2007). Conduct problem subtypes in males predict differential adult health burden. Archives of General Psychiatry, 64, 476-484.

Odgers, C. L., Moffitt, T. E., Broadbent, J. M., Dickson, N., Hancox, R. J., Harrington, H., et al. (2008). Female and male antisocial trajectories: From childhood origins to adult outcomes. Development and Psychopathology, 20, 673-716.

Raine, A., Moffitt, T. E., Caspi, A., Loeber, R., Stouthamer-Loeber, M., & Lynam, D. R. (2005). Neurocognitive impairments in boys on the life-course persistent antisocial path. Journal of Abnormal Psychology, 114, 38-49.

Robins L. N. (1966). Deviant children grown up. Baltimore: Williams & Wilkins.

Trouton, A., Spinath, F. M., & Plomin, R. (2002). Twins Early Development Study (TEDS): A multivariate, longitudinal genetic investigation of language, cognition and behavior problems in childhood. Twin Research, 5(5), 444-448.

Our group has already completed a three-stage genomewide association study of mathematical ability in 10 year-olds drawn from the Twins Early Development Study (TEDS) (under review). 43 SNPs were nominated from the first two stages which involved assaying pooled DNA on Affymetrix DNA microarrays to compare individuals of high mathematical ability to those of low mathematical ability. In the third stage, these 43 SNPs were individually genotyped in a sample which spanned the entire spectrum of ability. 10 SNP associations remained significant. Though effect sizes were extremely small, when a composite SNP-score was created for subjects in the final sample by summing the number of 'mathemtics score increasing alleles', these 10 SNPs accounted for 2.9% of the phenotypic variation in the sample. Interestingly, when the remaining 33 SNPs were added to this score, they did not mask this effect on mathematical performance, but rather increased it. This suggests to us that many of these remaining 33 SNPs may show true influences over mathematical ability which when analysed alone, our sample was underpowered to detect.

As the true test of association is replication, we are currently seeking out a comparable sample with mathematics ability data. With access to 1) existing ALSPAC mathematics scores, and 2) existing genotype data - collected on Illumina microarrays - we would be able to simply and easily test for the association of the 43 implicated SNPs (or where necessary, their proxies on the Illumina platform) with mathematical ability in the ALSPAC sample. As quantitative genetic results indicate that some of the genetic action over mathematics is general across ages, whilst some is not, we propose to investigate the action of the 43 SNPs over mathematical ability at all ages for which there is data available (we have been informed that this is 4-5, 11 and 15 years). Further to this, we are especially interested in investigating the combined influence of 1) the top 10 SNPs and 2) all 43 SNPs over mathematical ability. As mentioned above, this may be achieved by simply summing the math-increasing alleles possesed by each subject, and investigating the relationship of this SNP-set score with mathematical ability. If linear models are used to investigate SNP associations, then a simple alternative might be to add all 10/43 SNPs into the model. Whilst we would greatly prefer to conduct these analyses ourselves - we do understand that there may be restrictions on the release of some or all of the data we have requested, and so would also be very happy to work closely with ALSPAC researchers on this project.

As much of the work our group conducts is funded by the Wellcome Trust, we follow the same open access publishing policy as ALSPAC. We will therefore ensure that any publications resulting from this collaboration will be freely available to all researchers via PubMed Central.

ALSPAC is an important resource in the study of child development. Researchers using the cohort have shown an association between factors such as antenatal diet, maternal and paternal anxiety and depression and later child development.(1-5) There is strong evidence that child development and school performance are also influenced by socioeconomic factors including: parental occupation, marital status, income and parental education. (6-10) Maternal smoking and social adversity, specifically during pregnancy, were independently associated with an increase in risk of child symptoms of inattention and hyperactivity in a large multi-centre study of environmental risk factors for ADHD.(11) However, the causal mechanism of how childhood adversity affects development is unclear.

Early parent-child interaction is the basis of the attachment relationship which in turn allows the child to develop a template for social and emotional relationships.(12) Although there has been work looking at the effect of parenting and childhood attachment difficulties on general health, behaviour and self regulation, potentially mediated through the hypothalamic-pituitary-adrenal axis, the evidence base for the effect of parenting on other aspects of childhood development and later scholastic competence is less clear.(13, 14) The NICHD Study of Early Child Care and Youth Development, looking at 1000 US families, showed that early attachment predicted emotional development and language skills in younger children but not cognitive functioning; however the effect of attachment was dependent on socioeconomic background.(15, 16)

Dr Waylen and Professor Stewart-Brown have worked extensively with ALSPAC and have developed and validated a parenting measure which incorporates parental reports of attitude, behaviours and feelings towards the child. This measure of parenting has been shown to be independently associated with maternal reported child health.(17) In addition, ALSPAC has repeated measures of self-reported parenting activity scores between 6 months and 38 months. Professor Emond, Dr Deave and Dr Lingam have experience in the analysis of the modified Denver Developmental Screening Test (modified DDST) measured at 6, 18, 30 and 42 months along with measures of later development including motor coordination, speech and language skills, non verbal skills, social communication, and academic ability (reading and spelling).(2, 18)

Research questions

1. What is the effect of parenting on child development measured using standardised tests, controlling for potential confounding and modifying factors?
2. To what extent does parenting mediate the effect of known risk factors for child development such as socioeconomic status and parental education?

Methods

The primary exposure variable will be the parenting measure developed by Dr Waylen and Professor Stewart-Brown. A secondary exposure variable Parenting Activity will be developed using the ALSPAC parenting activity scores.

Outcome variables will be the modified Denver Developmental Screening Test (modified DDST) measured at 6, 18 and 30 months and later developmental measures including inattention and hyperactivity using the Development and Well-Being Assessment- DAWBA; difficulties in social cognition (The Social and Communication Disorders Checklist- SCDC; poor reading skills (using the Weschler Objective Reading Dimensions- WORD and NARA II); expressive language and comprehension using the WOLD, nonverbal impairment using the DANVA and Cognitive ability using the abbreviated WISK III measured between 7 and 9 years. (19-25)

A series of multivariable linear regression and logistic regression models (for continuous and binary outcomes respectively) will be created to look at the associations between parenting measures and child development accounting for potential confounding and mediating factors identified from the literature. Adjustment will be made for the bias introduced by missing data using multiple imputation by chained equation.(26)

Variable needed

Dr Lingam already has a dataset containing the developmental variables and some of the parenting activity scores needed for the analysis as part of his PhD looking at the causes and consequences of developmental coordination disorder. Use of the derived parenting measure has been granted by Dr Waylen and Professor Stewart Brown. Full details of all the variables in the analysis are attached in table 1.

Power calculation

The prevalence of suboptimal parenting (increased hostility, resentment and hitting/shouting) has been reported in between 10 and 30% of the ALSPAC cohort at different time points.(17) Taking 7000 to be the study population (approximate children that attended focus at 7) and our outcome variable to be the lowest 10% of a given developmental trait, the relative risk that it will possible to be 80% sure of detecting at the 5% level of significance is between 1.27 and 1.42.

References

1. Hibbeln JR, Davis JM, Steer C, Emmett P, Rogers I, Williams C, et al. Maternal seafood consumption in pregnancy and neurodevelopmental outcomes in childhood (ALSPAC study): an observational cohort study. Lancet. 2007 Feb 17;369(9561):578-85.

2. Deave T, Heron J, Evans J, Emond A. The impact of maternal depression in pregnancy on early child development. BJOG. 2008 Jul;115(8):1043-51.

3. Ramchandani P, Stein A, Evans J, O'Connor TG. Paternal depression in the postnatal period and child development: a prospective population study. Lancet. 2005 Jun 25-Jul 1;365(9478):2201-5.

4. Ramchandani PG, Stein A, O'Connor TG, Heron J, Murray L, Evans J. Depression in men in the postnatal period and later child psychopathology: a population cohort study. J Am Acad Child Adolesc Psychiatry. 2008 Apr;47(4):390-8.

5. Talge NM, Neal C, Glover V. Antenatal maternal stress and long-term effects on child neurodevelopment: how and why? J Child Psychol Psychiatry. 2007 Mar-Apr;48(3-4):245-61.

6. Brooks-Gunn J, Duncan GJ. The effects of poverty on children. Future Child. 1997 Summer-Fall;7(2):55-71.

7. Duncan GJ, Brooks-Gunn J. Family poverty, welfare reform, and child development. Child Dev. 2000 Jan-Feb;71(1):188-96.

8. Votruba-Drzal E. Economic disparities in middle childhood development: does income matter? Dev Psychol. 2006 Nov;42(6):1154-67.

9. Prathanee B, Thinkhamrop B, Dechongkit S. Factors associated with specific language impairment and later language development during early life: a literature review. Clin Pediatr (Phila). 2007 Jan;46(1):22-9.

10. Stanton-Chapman TL, Chapman DA, Bainbridge NL, Scott KG. Identification of early risk factors for language impairment. Res Dev Disabil. 2002 Nov-Dec;23(6):390-405.

11. Rodriguez A, Olsen J, Kotimaa AJ, Kaakinen M, Moilanen I, Henriksen TB, et al. Is prenatal alcohol exposure related to inattention and hyperactivity symptoms in children? Disentangling the effects of social adversity. J Child Psychol Psychiatry. 2009 Feb 27.

12. Bowlby J. Attachment and loss: Attachment Vol1 London: Hogarth Press; 1969.

13. Rees CA. Thinking about children's attachments. Arch Dis Child. 2005 Oct;90(10):1058-65.

14. Stewart-Brown SL, Fletcher L, Wadsworth ME. Parent-child relationships and health problems in adulthood in three UK national birth cohort studies. Eur J Public Health. 2005 Dec;15(6):640-6.

15. Belsky J, Fearon RM. Infant-mother attachment security, contextual risk, and early development: a moderational analysis. Dev Psychopathol. 2002 Spring;14(2):293-310.

16. Friedman SL, Boyle DE. Attachment in US children experiencing nonmaternal care in the early 1990s. Attach Hum Dev. 2008 Sep;10(3):225-61.

17. Waylen A, Stallard N, Stewart-Brown S. Parenting and health in mid-childhood: a longitudinal study. Eur J Public Health. 2008 Jun;18(3):300-5.

18. Lingam R, Hunt L, Golding J, Jongmans M, Emond A. Prevalence of developmental coordination disorder using the DSM-IV at 7 years of age: a UK population-based study. Pediatrics. 2009 Apr;123(4):e693-700.

19. Frankenburg WK, Dodds J, Archer P. The Denver II: a major revision and restandardization of the Denver Developmental Screening Test Pediatrics. 1992;89 (1):91-7.

20. Goodman R, Ford T, Richards H, Gatward R, Meltzer H. The Development and Well-Being Assessment: description and initial validation of an integrated assessment of child and adolescent psychopathology. Journal of Child Psychology and Psychiatry. 2000 Jul;41(5):645-55.

21. Skuse DH, James RS, Bishop DV, Coppin B, Dalton P, Aamodt-Leeper G, et al. Evidence from Turner's syndrome of an imprinted X-linked locus affecting cognitive function. Nature. 1997 Jun 12;387(6634):705-8.

22. Rust J, Golombok S, Trickey G. WORD Wechsler Objective Reading Dimensions Manual. Sidcup, UK: The Psychological Corporation; 1993.

23. Rust J. WOLD Wechsler Objective Language Dimensions Manual. London, UK: The Psychological Corporation; 1996.

24. Nowicki S, Duke MP. Individual differences in the nonverbal communication of affect: the Diagnostic Analysis of NonVerbal Accuracy scale. Journal of Nonverbal Behaviour. 1994;18:9-35.

25. Weschsler D, Golombok S, Rust J. WISC-IIIUK Wechsler Intelligence Scale for Children -Third Edition UK Manual. 3rd ed. Sidcup, UK: The Psychological Corporation; 1992.

26. Stuart EA, Azur M, Frangakis C, Leaf P. Multiple imputation with large data sets: a case study of the Children's Mental Health Initiative. Am J Epidemiol. 2009 May 1;169(9):1133-9.