Proposal summaries
B1055 - Identifying a subgroup of oppositional children who are at risk of depression - 06/10/2010
Fewer than twenty years ago, developmental psychopathologists and psychiatrists believed that there was little overlap between externalizing childhood psychiatric disorders, such as oppositional defiant disorder (ODD), and internalizing disorders, such as depression. There is now ample evidence from my own research (Boylan et al., 2007, 2010) and research of others (Copeland et al., 2009; Nock et al,.2007; Burke et al., 2010; Stringaris and Goodman, 2009), that childhood ODD shows significant concurrent and prospective associations or comorbidity with later depression. This discovery is changing the way we treat youths with ODD, and think about the causes of the disorder.
Despite these advances, important questions remain unanswered: Do all youths with ODD develop depression or will this only occur in sub-groups? What are the distinguishing characteristics of children belonging in one or other subgroups? The answers to these questions have profound clinical implications. If all youths with ODD develop depression, then perhaps we should begin prophylactic depression treatment, even before symptoms occur. On the other hand, if we can identify sub-groups of youths with ODD that develop depression, then prevention and treatment protocols can be applied in a more sophisticated and effective manner.
Based on my earlier work and the works of others which have demonstrated ODD symptoms cluster in various ways in the population , I propose that there are sub-groups of youths with ODD that will develop depression over time and the ability to identify and validate these groups is of substantial importance.
Study Aims:
(1) To determine whether oppositional symptom sub-groupings identified in previous factor analytic studies can be linked to groups of children who are oppositional.
(2) To determine whether these groups show differential risk for developing depression over time. If there is not strong evidence for person-based groupings, we will test how oppositional behaviours are associated with risk of depression across childhood.
Research objectives:
(1) To assess whether oppositional symptom subgroups identified by others (ie. a predominantly negative emotion and a predominantly antisocial symptom type) can be reliably identified in individual children using cross sectional and longitudinal data from an epidemiologic cohort sample.
(2) To determine if one, or more, of these groups is differentially associated with depression in late childhood and adolescence.
(3) To assess whether membership in these groups can be predicted on the basis of neurobiological and psychosocial risk factors that are associated with the ability to regulate negative emotions, in each of the following domains:
The child themselves (physiology of stress response, early life temperament, cognitive abilities, aggression, anxiety comorbidity)
The family context (family functioning, early attachment, parent discipline, family socioeconomic and family structure, parent mental health)
The peer and social context: peer relationships, school connectedness, social skills
Proposed methods:
We propose to use previously-collected questionnaire data for children from the ALSPAC study. The main outcome variables are the ODD and major depression and dysthymia symptoms from the DAWBA (at ages 7, 10 & 16) as reported by parents. Additional covariables, predominantly from questionnaire data, will be requested to distinguish the ODD groups and or predict relationships between ODD symptoms and depression based on study hypotheses. Longitudinal data on the DAWBA responses for the entire cohort will be requested to understand the impact of sample loss over time as complete data on DAWBA assessment periods is not necessary to conduct the analyses.
Analytic strategy:
Objective 1: A two step approach will test for the presence of groups of children with i) different types of ODD symptoms and ii) the stability (or reliability) of these prototypes/groups over time. The clustering of ODD symptoms will be identified from the DAWBA ODD symptoms data using factor analysis, followed by a confirmatory method (longitudinal confirmatory factor analysis). In step 2, we will test whether these (?#) identified ODD groups are developmentally distinct in that they can be distinguished based on the course of their symptom trajectory over time. To do this, we will use the group-based trajectory approach (with the Proc Traj procedure in SAS) which estimates the number and shape (course) of longitudinal trajectory groups in the sample, the proportion of children in each group and a probability for each child to belong to each group. Trajectories will be estimated for each ODD symptom cluster that is identified. Following this step, an extension of Proc Traj allows trajectory groups of one symptom cluster to be linked probabilistically to groups of another symptom cluster as joint trajectories to describe how the course of "comorbid" symptoms potentially overlap within individual children. As many children are likely to have both clusters of ODD symptoms over time, but most will have one or the other cluster (ie not have symptom overlap), assessing joint trajectories will establish how common these "comorbid" and non-comorbid children are. If comorbidity between ODD clusters is common, this suggests that there are not distinct ODD subtypes or subgroups.
In summary, the outputs of these analyses will identify: 1) are there likely ODD clusters? 2) how common are they? 3) how distinct are they from each other in terms of developmental course or how commonly do they overlap?
Objective 2: The identified joint oppositional symptom trajectory groups will be used to predict youth self-report of depression as an outcome at age 16 (measured as the youth DAWBA) using an extension in Proc Traj for this purpose. The output here will consist of a regression coefficient describing the strength of association between the various identified joint trajectory with the outcome of depression.
Objective 3: The same joint trajectory groups will be compared for their differential association with covariables (based on hypotheses which we do not outline here given space issues) as a means of exploring how they may differ etiologically from each other and how they may be differentially associated with risk of depression. Logistic regression will be conducted to examine if the prevalence rate of each covariate differs across trajectory groups.
B1053 - A Genome-wide association of variation in cotinine levels within the ALSPAC study PhD - 02/10/2010
Evidence within the literature and other studies has suggested a role in the influence of genetics on adolescent tobacco smoke, and eventual addiction to that substance (Li 2003). Genome wide association studies (GWAS) have provided a very robust methodology for identification of causal genetic influences on complex disease phenotypes.
The abuse of tobacco in adolescence can cause addiction in future life increasing your chance to have a lower mortality age (Aklin et al. 2009). It is a very important societal problem as tobacco smoke has one of the highest death rates of a preventable death rates, tobacco use kills approximately 5-6 million people annually worldwide, accounting for 1 in 5 of all male deaths and 1 in 20 of all female deaths. It is predicted to kill 450 million adults, on the basis of current consumption, between 2000 and 2050 (Jha 2009). The Caporaso et al. (2009) genome wide association study into different smoking phenotypes, such as duration and age of smoking initiation, found no loci at genome-wide significance (pless than 10-7), although there were some suggestive hits. Therefore refinement of the phenotype to use a more biological measure may contribute more to our understanding of this disorder and increase of number of individuals as we would be able to use more of the ALSPAC sample.
Tobacco abuse contains the psychotropic chemical nicotine; cotinine is an alkaloid found in tobacco and is a metabolite of nicotine. This substance is a good proxy for amount of nicotine within the individual and therefore the amount smoked by the individual. This phenotype will not be prone to reporter bias; although the chemical is short-lived within the body therefore there could be some measurement error according to the smoking behaviour of the individual. The amount of this substance will be the principle phenotype used within this GWAS to investigate whether any possible signals of genetic loci at genome-wide significance (pless than 10-7) can be found. This can then be compared to other candidate gene studies to investigate if any of the genetic loci of biological relevance can be found within the signals, or if novel signals have been discovered for the amount of cotinine within an individual. The best measure of this would be at 15 + years of age, as this is around a similar time to when we have self reported measure of tobacco use, which can be compared as a control measure. Cotinine can be measured in the urine or saliva samples which ALSPAC receive from clinics.
The principle aim of this project is to discover novel genetic loci which may contribute toward the complex disorder of tobacco abuse, which may give us some early indications towards adolescent abuse to adult addiction. This project has a novel approach by using the quantitative biological sampling of cotinine, which has no reporter bias and is generally less biased than using questionnaire data which is common for previous studies.
For this project cotinine levels would be needed on all individuals where it has been measured, also the genetic information for ALSPAC which is currently available. Also covariates could be retrieved from the ALSPAC folder from previous questionnaires. We will undergo the GWAS for the continuous variable of cotinine using related programmes; the covariates would be measured for an association with the main phenotype and then added to the GWAS. Covariates could include parental smoking behaviours, parental smoking, socio-economic status and phenotypes related to smoking.
References
Aklin, W.M., Moolchan, E.T., Luckenbaugh, D.A., & Ernst, M. 2009. Early tobacco smoking in adolescents with externalizing disorders: Inferences for reward function. Nicotine Tobacco Research, 11, (6) 750-755 available from: http://ntr.oxfordjournals.org/cgi/content/abstract/11/6/750
Jha, P. 2009. Avoidable global cancer deaths and total deaths from smoking. Nat Rev Cancer, 9, (9) 655-664 available from: http://dx.doi.org/10.1038/nrc2703
Li, M.D. 2003. The Genetics of Smoking Related Behavior: A Brief Review. The American Journal of the Medical Sciences, 326, (4) available from: http://journals.lww.com/amjmedsci/Fulltext/2003/10000/The_Genetics_of_Smoking_Related_Behavior__A_Brief.3.aspx
B1052 - Maternal CHRNA3 genotype prenatal smoking and maternal blood pressure in pregnancy - 02/10/2010
Pre-eclampsia is a syndrome of pregnancy that is marked by proteinuria and hypertension. It is a leading cause of maternal mortality and is also associated with fetal growth restriction, placental abruption and perinatal death 1. Interestingly, there is a well-documented association between maternal smoking during pregnancy and reduced risk of pre-eclampsia. A recent review of epidemiological studies reported a total of 48 studies, consistently reporting a reduction in risk of pre-eclampsia in maternal smokers - up to 50% reduction in heavy smokers 2. However, ascertaining with certainty the causality of this association remains a challenge. Whilst the association has long been observed in the literature, the biological mechanisms for this relationship remain unknown. Indeed, many of the known effects of smoking suggest that prenatal smoking should increase the risk of pre-eclampsia, rather than decrease it.
Furthermore, whilst the association between maternal smoking and pre-eclampsia is well documented, associations with antenatal blood pressure per se are less well established. Few studies have reported on this, with some suggestion that smoking may be associated with increased systolic blood pressure3;4, in constrast to what might be expected from the observed associations with pre-eclampsia. Several of the proposed authors for this project (MacDonald-Wallis; Lawlor) have been analysing trajectories of blood pressure in pregnancy in the ALSPAC mothers, and have observed systematic differences in the trajectories of blood pressure amongst women who were nonsmokers, quitters and continued smokers during pregnancy (manuscript in progress). We would like to extend this work by analysing these trajectories with respect to a maternal genotype known to be associated with smoking during pregnancy, in order to better ascertain the extent of the causal mechanisms involved.
Research Questions:
1) Is maternal prenatal smoking status (as indexed by maternal CHRNA3 genotype) causally related to risk of pre-eclampsia
2)Is maternal prenatal smoking status (as indexed by maternal CHRNA3 genotype) causally related to trajectories of blood pressure during pregnancy.
(1) Sibai B, Dekker G, Kupferminc M. Pre-eclampsia. Lancet 2005; 365(9461):785-799.
(2) England L, Zhang J. Smoking and risk of preeclampsia: a systematic review. Front Biosci 2007; 12:2471-2483.
(3) Bakker R, Steegers EAP, Mackenbach JP, Hofman A, Jaddoe VWV. Maternal smoking and blood pressure in different trimesters of pregnancy: The Generation R Study. Journal of Hypertension 2010; 28.
(4) Matkin CC, Britton J, Samuels S, Esenazi B. Smoking and blood pressure patterns in normotensive pregnant women. Paediatr Perinat Epidemiol 1999; 13:22-34.
B1054 - GWAS of Ponderal Index at Birth - 29/09/2010
This project involves assessing the genetic contribution to the determination of body composition at birth. Fetal and early postnatal life has been well demonstrated as a period of metabolic "programming" with the potential to influence a wide of adult diseases. Clearly there is contribution from, and interaction between, various genetic and environmental influences to determine and individual's outcome in the face of the early life environment.
Early work in this field first identified birth weight as an indicator of growth during pregnancy, but subsequent work has shown that birth weight is a somewhat crude measure, and more information can be gained by using more refined assessments of early life growth. Ponderal index is a measure of body composition that has been well validated at birth. It is analogous to the adult body mass index (with body length cubed, rather than squared as in BMI) compensating for body mass compared to length (height), and is well correlated to other measures of adiposity. Compared with birth weight, ponderal index is better at identifying "asymmetrically" growth restricted neonates in whom we suspect metabolic stressors to be greater than constitutionally small babies.
This project is a collaborative approach among members of the Early Growth Genetics Consortium. Cohorts with appropriate genetic data will use a genome-wide approach to identify genetic loci associated with changes in ponderal index. Each cohort will run individual analyses according to a uniform multivariate model, and then meta-analysis will look at all of the cohorts in combination. Attempts at replication of significant findings will be made in other birth cohorts without complete GWAS data. With an anticipated total number exceeding 10,000 individuals, this study will be well powered to show the relatively small changes associated with single genetic variants. This will be the first GWAS of ponderal index to be performed.
B1050 - Comparisons of dietary patterns and health outcomes in the ALSPAC and Raine studies - 28/09/2010
Background
The causes of obesity in children are complex and currently not well understood. However, dietary intake, physical activity, socio-economic status, genetic influences and early social factors are thought to be important. Despite the increasing prevalence of obesity in children, few longitudinal studies are available to clarify these relationships. We are interested in dietary patterns that may lead to obesity and the factors influencing these dietary patterns.
Aims
The aims of this study are to conduct cross-cohort comparisons between the Western Australian Pregnancy (Raine) Cohort Study and ALSPAC, which are contemporaneous cohorts. We wish to compare the role of dietary patterns in obesity and the early determinants of obesogenic dietary patterns in these cohorts.
The Raine Study is a pregnancy cohort study of 2900 pregnant women recruited between 16 and 20 weeks of gestation through the public antenatal clinic and local private clinics in Perth, Western Australia from May 1989 till November 1991 (Newnham, et al, 1993 ). A total of 2868 children have been followed up at birth and ages 1, 2, 3, 5, 8, 10, 14 and 17 years. This study will be focusing on children at 14 and 17 years of age.
Specifically, we propose to:
1) Compare dietary patterns identified using reduced rank regression (RRR) in the Raine and ALSPAC cohorts
2) Test whether dietar patterns in the Raine study can predict obesity in ALSPAC children.
3) Identify and compare early life determinants of dietary patterns in ALSPAC and Raine cohorts. The early life determinants will include family structure and maternal education, parental pre-pregnancy body weight and BMI, maternal working hours, maternal smoking status during pregnancy and paternal smoking status. Identifying the early determinants of dietary patterns will help to identify target areas and developmental periods for improving dietary patterns.
Methods
1) Identify a dietary pattern using reduced rank regression (RRR) in the Raine cohort at 14 years of age, that explains the most variation in:
(i) dietary energy density (DED)
(ii) fibre density (FD)
(iii) percentage of total energy from fat
These particular dietary components are hypothesised to influence obesity risk.
2) Examine whether the dietary pattern identified in the Raine cohort predicts obesity in the Raine and ALSPAC cohorts. Using confirmatory RRR, we will apply the Raine dietary pattern identified at 14 years of age to ALSPAC food intakes collected at 13 years of age. Using linear regression and adjusting for appropriate confounders, we will examine prospective associations between applied dietary pattern scores at 13 y of age and obesity at 15 y of age in the ALSPAC cohort.
3) Identify early maternal and socio-demographic determinants of the Raine dietary pattern: analysis of variance (ANOVA) will be used to examine the variation in dietary pattern scores that is explained by early factors such as pre pregnancy BMI, maternal smoking status, maternal education, and maternal working hours. We propose to undertake the same analysis in the ALSPAC cohort to be able to compare the early determinants of the same dietary pattern in the Raine and ALSPAC cohorts.
ALSPAC data required for this project:
Food and nutrient intakes:
3-day food diaries completed at 13 years of age or FFQ data ?
Derived dietary intakes :
Cambridge food groups;
dietary energy density;
fibre density;
% energy from fat;
total energy intake;
misreporting of dietary intake (using Torun's method).
Anthropometry:
body mass index;
waist circumference;
fat mass;
pubertal stage;
birth weight and length;
gestational age
Physical activity:
accelerometer counts incl. moderate-vigorous activity
Socio-demographics:
child age;
child gender;
maternal age at birth;
maternal education;
maternal pre-pregnancy BMI;
maternal smoking status (pre pregnancy and up to 5 years later);
paternal smoking status;
maternal occupation and working hours;
family structure;
family functioning
Reference:
Newnham JP, Evans SF, Michael CA, Stanley JF, Landau LI. Effects of frequent ultrasound during pregnancy - a randomised controlled trial. Lancet 1993;342:887-91.
B1051 - Maternal Iron Genotype Blood Pressure Trajectories in Offspring and During Pregnancy and Maternal Gestational Diabetes - 28/09/2010
Background
Maternal anemia during pregnancy has previously been reported to be associated with elevated offspring blood pressure in later life1;2. However, this has only been reported in a small number of studies, with others failing to replicate this association3-5. Indeed, we previously explored this in ALSPAC6 and did not observe any consistent relationship between indicators of maternal iron status and offspring blood pressure at 7 years. A potential limitation of such studies is that assessing maternal iron status from haemoglobin measures is complicated by the tendency for low antenatal haemoglobin levels to reflect the natural haemodilution of pregnancy, rather than being indicative of pathological iron deficiency anemia. As such, there may be limited ability to identify causal effects of maternal iron deficiency anemia within such studies and this may have contributed to the inconsistent associations previously observed. We would like to build on our previous work carried out in ALSPAC by a) exploring maternal genetic variants affecting iron status as proxies for maternal iron, in order to better identify causal effects of maternal iron status and b) using trajectories of blood pressure across childhood and adolescence as the outcome measure, rather than one single measure of blood pressure. This will allow us to examine whether maternal iron status influences blood pressure across childhood or only at specific ages.
In contrast to the hypothesised adverse effects of maternal anemia during pregnancy, some have argued that iron depletion during pregnancy might in fact represent an adaptive physiological condition to prevent the adverse effects of oxidation, insulin resistance and thrombosis7. Indeed, whether iron supplementation in pregnancy is necessary, or even toxic, is still controversial and there is some evidence that maternal iron supplementation in pregnancy is associated with adverse outcomes such as increased risk of gestational diabetes and hypertensive disorders of pregnancy7;8. Thus, in addition to studying trajectories of child blood pressure, we would also like to assess the relationship between genetic variation in maternal iron status and trajectories of maternal blood pressure during pregnancy and gestational diabetes/glycosuria in ALSPAC.
Research Questions:
1) Is maternal iron status in pregnancy (as indicated by maternal iron genotypes) causally associated with offspring baseline BP at age 7 and BP trajectories between ages 7 and 15 years?
2) Is maternal iron status in pregnancy (as indicated by maternal iron genotypes) causally associated with maternal BP trajectories in pregnancy? For these analyses women with existing hypertension will be excluded.
3) Is maternal iron status in pregnancy (as indicated by maternal iron genotypes) causally associated with maternal gestational diabetes/glycosuria? This combined outcome will be used because there are too few cases of gestational diabetes alone for meaningful analyses; the use of glycosuria is justified by previous research in ALSPAC showing that it is associated with macrosomia9. For these analyses mothers with existing diabetes will be excluded.
Reference List
(1) Law CM, Barker DJ, Bull AR, Osmond C. Maternal and fetal influences on blood pressure. Arch Dis Child 1991; 66(11):1291-1295.
(2) Godfrey KM, Forrester T, Barker DJ, Jackson AA, Landman JP, Hall JS et al. Maternal nutritional status in pregnancy and blood pressure in childhood. Br J Obstet Gynaecol 1994; 101(5):398-403.
(3) Whincup P, Cook D, Papacosta O, Walker M, Perry I. Maternal factors and development of cardiovascular risk: evidence from a study of blood pressure in children. J Hum Hypertens 1994; 8(5):337-343.
(4) Bergel E, Haelterman E, Belizan J, Villar J, Carroli G. Perinatal factors associated with blood pressure during childhood. Am J Epidemiol 2000; 151(6):594-601.
(5) Belfort MB, Rifas-Shiman SL, Rich-Edwards JW, Kleinman KP, Oken E, Gillman MW. Maternal iron intake and iron status during pregnancy and child blood pressure at age 3 years. Int J Epidemiol 2008; 37(2):301-308.
(6) Brion MJ, Leary SD, Davey Smith G, McArdle HJ, Ness AR. Maternal anemia, iron intake in pregnancy, and offspring blood pressure in the Avon Longitudinal Study of Parents and Children. Am J Clin Nutr 2008; 88(4):1126-1133.
(7) Bo S, Menato G, Villois P, Gambino R, Cassader M, Cotrino I et al. Iron supplementation and gestational diabetes in midpregnancy. Am J Obstet Gynecol 2009; 201(2):158-6.
(8) Ziaei S, Norrozi M, Faghihzadeh S, Jafarbegloo E. A randomised placebo-controlled trial to determine the effect of iron supplementation on pregnancy outcome in pregnant women with haemoglobin greater than or = 13.2 g/dl. BJOG 2007; 114(6):684-688.
(9) Lawlor DA, Fraser A, Lindsay RS, Ness A, Dabelea D, Catalano P et al. Association of existing diabetes, gestational diabetes and glycosuria in pregnancy with macrosomia and offspring body mass index, waist and fat mass in later childhood: findings from a prospective pregnancy cohort. Diabetologia 2010; 53(1):89-97.
B1047 - Epigenetic changes in the development of obesity and associated metabolic disorders - 27/09/2010
Obesity and its related metabolic disorders represent a major social, economic and health burden.
There is immense interest in epigenetic processes and the role that they might play in mediating
complex disease risk through their influence on gene regulation. This project adopts an
epidemiological approach, including the development of novel data analysis methods, and aims to
further our understanding of the contribution made by epigenetic mechanisms to obesity and its
sequelae, including non alcoholic fatty liver disease.
This study will measure DNA methylation patterns in children at birth and at age 15 and relate
these to a wide array of exposures and obesity-related traits, and importantly, the trajectory of
these traits. It represents by far the largest study of DNA methylation and its association with
obesity-related exposures (beginning in utero) and outcomes to date. The study will utilise a world
leading longitudinal study which has followed children from birth to age 17 (with future follow-up
planned) and has amassed an unprecedented amount of data on these individuals. To
complement large, well-powered human epidemiological studies of peripheral blood DNA
methylation, analysis of tissue specific DNA methylation will be analysed in human biopsy samples
and both DNA methylation and gene expression will be analysed in a range of target tissues in
animal models.
The study will identify obesity-related DNA methylation patterns using the Illumina 450k human
methylation array in 250 children at birth and age 15. From these data a custom panel of
methylation sites associated with obesity-related exposures and the phenotype itself will be
designed. A total of 1,200 children will then be analysed to establish the relationship between
methylation variation and obesity and its associated metabolic disorders. In addition the
association of DNA methylation signatures with common genetic variation will be investigated and
data analysis techniques developed to strengthen causal inference. Finally tissue specific
methylation patterns will be explored using human liver tissue and multiple tissues from animal
models of obesity.
The study will make a significant contribution to the emerging field of epigenetic epidemiology and
the data arising will be made available to the wider scientific community.
B1046 - Functional mapping of Quantitative Trait Loci underlying childrens growth - 14/09/2010
Many traits, such as body height and body weight, undergo a developmental change. Traditional genetic analyses of these traits are based on the association between the genotype and phenotype measured at individual times. But this approach does not consider the dynamic feature of the traits and, therefore, limits the scope of its inference about genetic control. A natural way is to connect the genotype with the dynamic trajectories of a trait, enabling geneticists and clinicians to study the temporal pattern of genetic control.
Such a dynamic model has been developed in our group at Penn State University. This model 1 integrates mathematical aspects of trait formation and progression with genetic variation within the framework of genome-wide association studies (GWAS). It thus shows substantial biological relevance of the results and, meanwhile, possesses significant statistical merits because of parsimonious modeling of the time-dependent genetic effects and longitudinal covariance structure. We have performed extensive simulation studies to test and validate the usefulness of this new model. Although favorable statistical properties have been detected, we have a keen interest in applying this model to a large-scale longitudinal data set. By reading your article published in PLoS Medicine 2 and looking at your webpage, we are very confident that our model matches your data extremely well. If we get your data, we will commit our time to analyze and interpret them, and write papers for publication jointly.
Our specific data needs are:
(1) genotype data for all the SNPs you may have in cohort studies.
(2) phenotypic data measured at multiple time points. In your case, body height and weight from birth to ages 11 years were measured. We are also interested in other longitudinal measures, such as blood pressure, but we will request these data later.
Our plan: Once we get these data, we will analyze them in a couple of weeks and report the results to you at various stages. Meanwhile, we will be drafting a manuscript together with your group and publish it jointly in a top journal. In particular, we will be collaborating closely throughout this project with Nicholas Timpson from the Department of Social Medicine, University of Bristol.
As to the authorship, the following is our suggestion: If the results are from reanalyzing your data with our new model, we suggest that we are the first author and you are coauthors. A leader from each group (Penn State and Bristol) can be jointly a corresponding author. If your data are new and we simply help you analyze, you should be the first and corresponding authors, but we request the coauthorship.
B1044 - A Genome-wide association study for the use of tobacco using cotinine levels - 14/09/2010
Not available
B1043 - What are the pre and postnatal determinants of volumetric bone density and geometry in the child from 8 to 17 years - 14/09/2010
AIM: To model the pre and postnatal determinants of volumetric bone density and geometry in the child at 8 to 17 years.
BACKGROUND
Osteoporotic fractures constitute a huge public health burden, with a currently estimated annual cost of 1.7 billion in the UK(1). The peak bone mass achieved in early adulthood is a major determinant of osteoporosis risk in older age(2). Peak bone mass is partly inherited, but the nature of the inheritance from either parent in terms of site specificity, size, geometry and volumetric density is uncertain. Additionally, evidence is accruing that environmental factors early in life may have long term influences on childhood bone development to peak(3). Thus poor growth in early life is associated with reduced bone mineral content and increased risk of hip fracture in adulthood(4;5). Maternal lifestyle, body build, physical activity, diet and circulating 25(OH)-vitamin D status during pregnancy are associated with offspring bone mass(6;7). Animal studies have demonstrated similar effects(8;9) and suggested epigenetic processes such as DNA methylation as potential mechanistic explanations for these gene-environment interactions(10-12). However, in humans, it is not clear whether influences on bone mass in the neonate retain a persisting effect through childhood, and secondly whether these effects are mediated through bone size, geometry or true volumetric density. Existing data from mother-offspring cohorts have mainly utilised DXA(13;14), which, as a two-dimensional construct, does not allow measurement of true volumetric density, or differentiation of effects on cortical vs trabecular bone. Maternal behaviour before and during pregnancy is likely to influence how she feeds and looks after her child. In order to elucidate a model of pre and postnatal genetic and environmental determinants of childhood bone size, density and geometry, it is therefore necessary to address several issues:
1) Inheritance (genetic) of bone size, geometry and density from mother and father.
2) Environmental factors acting before and during pregnancy (e.g. maternal diet) which might act on fetal bone development in utero and potentially alter the long term growth trajectory.
3) Environmental factors observed before and during pregnancy which merely act as markers of later behaviour and child rearing (e.g. an influence of maternal diet in pregnancy on offspring bone development could act through an association between maternal diet in pregnancy and that of the child postnatally).
4) Postnatal environmental factors which may have temporary or permanent effects on bone (e.g. physical activity and calcium/ vitamin D exposure).
In this proposal we aim to use two large, well established mother-offspring cohorts to attempt to address these problems (SWS and ALSPAC).
Summary of Southampton Women's Survey (SWS): The SWS(15) is a unique ongoing prospective population cohort 12,583 non-pregnant women aged 20-34 years, resident in the city of Southampton and recruited through their general practitioners. Each woman was visited at home for an interview, which involved a questionnaire enquiring about diet, employment, physical activity, general health, alcohol consumption and cigarette smoking. Nutrition was assessed from a previously validated 100-item food frequency questionnaire. Detailed anthropometric measurements of the women (height, weight, leg length, skinfold thickness at the triceps, biceps, subscapular and supra-iliac regions) were made in order to assess body composition. From these 12,583 women, 3156 became pregnant and delivered infants within three years of initial interview. At early (11 weeks) and late (34 weeks) pregnancy further assessments were made. 1003 of these neonates underwent DXA of their whole body and lumbar spine within two weeks of delivery. Subsequent follow up at 4 (n=900, complete) and 6 (n=600, ongoing) years has included DXA (Hologic Discovery: whole body, lumbar spine and hip) and objective physical activity (combined accelerometer and heart rate monitor: Actiheart, Cambridge Neurotechnology, Cambridge, UK) assessment over 7 days continuously. Simultaneous recording of maternal physical activity by Actiheart is available at 4 years.
METHODS
The SWS 8-year follow-up is underway. Children are invited to attend for whole body, lumbar spine and hip DXA measurements (Hologic Discovery instrument). Anthropometry and grip strength measurements will also be made and a lifestyle/dietary questionnaire completed.
In order to model the influence of parental bone mass, and demarcate relationships by bone size, geometry and true volumetric density, we now propose to assess the parents of the SWS children, in addition to further refining the bone phenotype of the offspring. Interested parents will be invited to attend a research clinic at the Osteoporosis Centre at Southampton General Hospital. Here, a lifestyle questionnaire will be administered; height, weight, skinfold thickness and grip strength will then be recorded. Bone mass will be assessed in two ways: A DXA scan (Hologic Discovery) will be performed and measurements made of whole body, lumbar spine and hip bone area, bone mineral content and areal BMD in addition to total and regional lean and fat mass. Peripheral QCT at the tibia (4 sites to assess areas of predominantly cortical and trabecular bone) using a Stratec XCT 2000 machine. Measurements will be made of cortical and trabecular volumetric density, cortical thickness and cross sectional area, periosteal and endosteal circumference, bending strength and fracture load in the x and y axis. Muscle mass measurements will also be made. At this visit the 8 year old child will also undergo the same pQCT assessment yielding 500 mother-father-child triplets with DXA and pQCT indices generated on the same instruments. In a subset of 250 triplets, an Actiheart monitor will be fitted to each of mother, father and child, to be worn continuously for 7 days and posted back in a padded reply paid envelope at the end of this period.
ANALYSIS
1) SWS: All data will be checked for validity and outliers. Variables will be assessed for normality and transformed in necessary. The relationships between the exposures and the bone outcomes will be initially explored in univariate models. Associations found to be statistically significant (pless than 0.05) will be included in multivariate models to generate the set of maternal, paternal and childhood variables which best explain the bone outcomes at 8 years. Collinearity between parental and childhood environment, e.g. diet, will also be investigated to distinguish between true early life effects and inherited environmental factors.
2) ALSPAC: This will utilise existing data which will already be have been cleaned. Variables will be assessed for normality and transformed in necessary. The analysis will comprise two phases:
The initial analysis will again employ univariate models to explore the associations between parental and childhood factors and childhood bone size, areal density and size corrected density at 9 years. A multivariate model will then be generated (incorporating a variables where pless than 0.05 for the relationship with child bone outcomes) to compare with that derived from the SWS.
The second phase will include a similar process of univariate then multivariate modelling with qQCT derived indices of tibial size, geometry and volumetric density at 15 and 17 years. The aim of this analysis will be to confirm whether influences observed at 8 years in SWS persist after puberty.
B1042 - Genes and happiness is there scope for Mendelian randomisation in the study of how well-being affects outcomes - 14/09/2010
The proposed research will build on research that attempts to replicate the association found between the serotonin transporter gene (SLC6A4) - specifically variation in the transcriptional control region (the 5-HTT-linked polymorphic region (5-HTTLPR)) - and well-being.
If an association is found, this research proposes to investigate whether Mendelian randomisation can be applied to study the effects of well-being on educational outcomes and social networks. There are a number of considerations to take into account before using genetic markers as instruments (see inter alia Lawlor et al. 2007[1], Lawlor et al. 2008[2], von Hinke Kessler Scholder et al. 2010[3]). An in depth study of whether the 5-HTTLPR and rs25531 can plausibly be used as instruments for well-being is warranted given an increasing interest in genes in economics literature.
While many studies investigate the determinants of well-being, positing that income, health and relationships can explain why some people are happier than others but Lyubomirsky et al. (2005)[4] argue that well-being is important in shaping outcomes - happier people are more successful at cultivating personal relationships and do better in the workplace. However, the bulk of evidence suggesting that well-being affects outcomes comes from cross-section or longitudinal studies and experimental studies. Cross-sectional studies document whether a correlation exists but not the direction of causality. On the other hand, if well-being is measured prior to the outcome, a more convincing case can be made that well-being caused the outcome, which is an advantage of longitudinal data. But this approach falters if well-being is highly correlated over time and it says little about what is driving changes in lagged well-being and therefore if lagged well-being is truly exogenous. Experimental studies provide a gold standard for studying the direction of causation, but by design these studies focus only on short-term changes in well-being. A suitable instrument for well-being would shed light on how long-term well-being affects outcomes and therefore assessing whether the 5-HTTLPR and rs25531 can be used as instruments for well-being, focusing on two different applications is a worthwhile exercise.
The focus on educational outcomes is motivated by the fact that educational outcomes are linked to job prospects and earnings and are therefore informative of the likely impact of well-being on economic circumstances. However, it is also an outcome where the exclusion restriction required for instrumental variable estimation may fail. Frisch et al. (2005)[5] previously show that well-being can predict academic retention. This research will examine whether the 5-HTTLPR and rs25531 are potentially useful instruments in an application to national exam scores.
The formation of social networks is an interesting area to verify whether mendelian randomisation is feasible because previous research on homophily shows that people with similar characteristics flock together (McPherson et al. 2001)[6] and research cited in Lyubomirsky et al. (2005)suggests that happier people are more appealing to others. The research will therefore examine whether this approach can be used to answer questions such as are people with similar levels of well-being more likely to befriend each other, does well-being influence the number of friends one has and position in friendship networks?
If an association is not found, the proposed research would investigate whether there is a gene-environment interaction using plausibly exogenous life stresses such as a death in the family.
B1040 - Obesity and obesity-related hypertension early genetic risk factors - 13/09/2010
1. RATIONALE
The proposed work is aimed at identifying genes contributing to blood pressure (BP) elevation related to excess body-fat in adolescence (i.e. preclinical obesity-induced hypertension). It builds on our previous research and is motivated by the following three overarching themes:
1.1 Adolescence is a period of preclinical disease: Adolescence is a period of human development when adult body composition and physiology develop and initial stages of cardiovascular disease, including hypertension, emerge1. Importantly for public health, this early preclinical stage may also be a period when disease trajectories can still be altered and full-blown diseases prevented.
1.2 Obesity is the leading risk factor for hypertension: Obesity has become a major health problem worldwide due to its growing prevalence and causal relationship to common chronic diseases, such as hypertension. It is spreading across all age groups, including children and adolescents. Epidemiological studies suggest that 65-78% of hypertension is due to obesity and that obesity-induced hypertension, as compared with obesity-unrelated hypertension, is associated with a greater risk for cardiovascular disease6. Despite the high public-health importance, our understanding of how excess body fat increases BP is still limited. We do know that the underlying mechanisms relate to not only excess but also intra-abdominal deposition of adipose tissue, and they commonly involve central sympathetic activation, insulin resistance, oxidative stress and augmented activity of the renin-angiotensin-aldosterone system.
1.3 Identification of genetic risk factors is a tool to uncover mechanisms of obesity-induced hypertension: The causal role of environment in the current epidemic of obesity and obesity-induced co-morbidities is undeniable. Genes, however, play a critical role in that they modify individuals' susceptibility to these disorders by regulating underlying mechanistic pathways. Therefore, identifying these genes may help us to uncover the pathways and thus open new avenues for prevention and treatment.
2. AIMS
Recent advances in high-through-put genotyping made significant inroads vis-a-vis the genetic architecture of hypertension, but preclinical stages and obesity-related hypertension have not been studied. To fill this gap, we propose to identify genes (and the underlying mechanisms) of preclinical stages of obesity-induced hypertension and to examine them across different ages. We propose to achieve this overarching goal by realizing Aims 1 and 2:
Aim 1: To identify genes and underlying mechanisms of adiposity and related BP variations in adolescence, we will perform genome-wide association studies (GWAS) with the Illumina Human610W-Quad BeadChip containing probes for greater than 550K single nucleotide polymorphisms (SNPs) in the Saguenay Youth Study (SYS). The SYS is an ongoing population-based investigation of 1,000 adolescents recruited from a French-Canadian founder population who undergo detailed cardiovascular, metabolic and behavioral phenotyping (15-hour protocol) . The SYS is ideally suited for the proposed studies for the following reasons: (1) it is an adolescent cohort allowing us to investigate preclinical stages of obesity-induced hypertension; (2) all adolescents are recruited from a genetic founder population where power of genetic analyses is expected to be higher (compared with regular outbred populations) due to more homogenous genetic background (and environment) and, hence, fewer genes contributing to the expression of complex traits, such as adiposity and BP; and (3) all adolescents undergo extensive high-fidelity phenotyping not only of adiposity and BP but also of variables that may provide additional mechanistic insight (e.g. sympathetic modulation of vasomotor tone) and characterize environment (diet, physical activity and psychosocial environment). To our knowledge, the SYS is currently the largest sample of adolescents (n=1,000) in whom intra-abdominal, visceral fat is measured directly with magnetic resonance imaging (MRI) and in whom sympathetic modulation of vasomotor tone is assessed with power spectral analysis of beat-to-beat BP. Furthermore, as we study predominantly disease-free adolescents (not yet hypertensive and still free of co-morbidities), this high-fidelity phenotyping is not confounded by e.g. BP-lowering medication. To date, we have phenotyped 750 adolescents and genotyped (610K SNPs) 600 of them, and we have published 19 peer-reviewed papers (with 2 additional being currently under review). We have also obtained encouraging results in our initial GWAS. The funds requested to meet Aim 1 are to support phenotyping of the remaining 250 adolescents, genotyping of the remaining 400 adolescents, as well as data management and performance of statistical analyses in all 1,000 adolescents. Note that we request funding for only cardiovascular and metabolic phenotyping of the remaining 250 adolescents, as the cost of their recruitment and both MRI and behavioral phenotyping is already covered by another CIHR grant (see Letter of Collaboration from Dr. Tomas Paus).
Aim 2: To replicate genotype-phenotype associations identified in Aim 1 across different ages (n=13,244 individuals). To achieve this aim we will test the associations in the SYS in the following replication cohorts: (1) the Avon Longitudinal Study of Parents and Children study (ALSPAC; 6,863 adolescents), which is a longitudinal birth cohort established in England (see Letter of Collaboration from Dr. George Davey-Smith); (2) the 1966 Northern Finland Birth Cohort (NFBC1966; 4,936 young adults), which is also a longitudinal birth cohort established in a Finish founder population (see Letter of Collaboration from Dr. Jarvelin Marjo-Riitta), and (3) the Quebec Hypertensive Family Study (QHFS; 897 older adults), which is a cross-sectional sample of adults recruited as members of hypertensive families from the same French-Canadian founder population as the SYS adolescents. All individuals to be analyzed in replication studies have already been phenotyped (BP and adiposity) and most of them were also genome-wide genotyped (317K SNPs in ALSPAC; 370K SNPs in NFBC1966). Therefore, the funds requested for Aim 2 are mainly to support data manipulation and management, and genetic analyses.
B1038 - Eating behaviour measurement using the mandometer A pilot study in ALSPAC - 10/09/2010
Not available
B1039 - Testing pathways between prenatal risks and child disruptive behaviour disorders - 08/09/2010
Attention deficit hyperactivity disorder (ADHD) and antisocial behaviour are common, extremely disabling disorders which have major adverse sequalae in childhood and later life. Despite being such an important clinical problem, their aetiology and pathogenesis are poorly understood. Available evidence suggests that both genetic and environmental risk factors are important.
Many studies have demonstrated significant association between prenatal exposure to alcohol, tobacco, marijuana and diabetes mellitus and child behaviour outcomes including ADHD and CD, although such findings are equivocal.
Each prenatal exposure has been shown by some animal studies to result in biological alterations to the developing foetus which could, in turn, influence child behaviour.
However, each of these prenatal risk factors is also associated with additional familial characteristics including social adversity, parental psychopathology and other social deficits, each of which have been associated with childhood behaviour problems. Furthermore, willingness to engage in risky behaviours during pregnancy is also associated with antisocial behaviour, impulsivity and ADHD symptoms in the mother.
Therefore, it is currently unknown whether these prenatal exposures lead to direct intrauterine effects on the developing foetus with long term consequences for childhood behavioural problems, or if observed associations arise instead via confounding inheritance of familial factors.
The aim of this proposal is to further explore the causal pathways for association between prenatal exposures and childhood ADHD and antisocial behaviour.
To this end, the aim of this proposal is to test the contrasting hypotheses that either:
1) Associations between prenatal exposure to alcohol, tobacco and marijuana, diabetes and childhood developmental disorders are the result of direct intrauterine effects OR
2) Such associations are due to genetic or other confounding factors.
The project will utilise data from three different samples, ALSPAC, the Cardiff IVF sample and data from the Swedish population registries.
Specifically in the ALSPAC sample, I propose to:
1. Test for associations between prenatal exposure to alcohol, tobacco, marijuana use, diabetes mellitus and offspring symptoms and diagnoses of ADHD and CD using regression analyses. Where this work has previously been undertaken, I will utilise previously published data or collaborate with other researchers.
2. Test which of the competing hypotheses account for these associations using methods comparing: Maternal and paternal exposure; pre and postnatal exposure; and a mendelian randomisation instrumental variable approach
3. Based on the findings of the earlier aims and suggestions from previous literature, I will investigate:
a. Whether causal associations arise via their influences on cognitive or other phenotypes or
b. Which confounding factors appear to best account for non-causal associations.
I propose to investigate genetic variants which are robustly associated with the risk exposures (smoking, alcohol and substance use and diabetes) which have been identified through GWAS and candidate gene studies. As this is a constantly changing field of research and much of the work is currently being undertaken by ALSPAC collaborators, I will nominate the proposed genetic variants at a later stage (of course with additional approval by the executive committee). However, it is anticipated that these will be variants previously genotyped within the ALSPAC sample, mainly on the Illumina 550k chip. I will also include funding in my proposal for additional genotyping of candidate variants which might not already be genotyped within the ALSPAC sample.
Current data required:
Parent & Teacher reports of ADHD & DAWBA diagnoses at 7 & 11 years
Parent & Teacher reports of antisocial behaviour & DAWBA CD diagnoses at 7 & 11 years
Child antisocial behaviour measures from focus visits
Age (parents & child at assessments)
Gender
Birth weight
Maternal & Paternal smoking during pregnancy
Current Maternal & Paternal smoking (current to ADHD diagnoses)
Mother & Father alcohol use during pregnancy and current (7 years)
Mother and father marijuana and other substance use during and after pregnancy
Mother and father diabetes diagnoses during pregnancy (maternal report and data from antenatal notes) and since birth
SES & Mother's education
Mother & Father psychopathology (time of pregnancy and current); ADHD, CD, Anxiety & Depression
Child cognitive assessments (IQ & neuropsychological test data, Key stage 1 & 2 results)
Mother & Child genotypes chosen from GWAS and association studies
B1037 - Maternal risky behaviour during pregnancy and child outcomes - 07/09/2010
This project has two aims:
1. To examine the causal effect of maternal smoking and drinking during pregnancy, and maternal BMI before pregnancy, on child development. More specifically, we are interested in outcomes relating to the child's birth, such as birth weight, as well as measures later in life such as academic outcomes, behavioural problems and child health (e.g. blood pressure).
2. To examine the effect of smoking and drinking on BMI or fat mass, both at the mother and child level.
As the relationships between (maternal) smoking, drinking, BMI and (child) development are likely to be subject to confounding in observational studies, we will identify the causal effect by using specific genetic markers that have been robustly identified to affect smoking, drinking or BMI as instrumental variables in a Mendelian randomization set-up.
We are aware that other researchers are also currently working on similar research questions using the ALSPAC data. For example, Sarah Lewis is involved in a project on alcohol led by Ron Gray, and we are aware that Rachel Freathy is looking at smoking-related variants. Hence, our aim is to explore this research question from a slightly different angle than that used in these existing projects. First, rather than examining the effect of these risky behaviours at the mean, we are specifically interested in their effects at different points in the outcome distribution. We will explore this distributional aspect with an instrumental variable quantile regression (IVQR) approach (Chernozhukov and Hansen, 2005, 2006). This will allow us to examine whether smoking and drinking has differential effects at different points in the (e.g.) birth weight or academic attainment distribution. Second, we are interested in the effect of different timings of smoking and drinking. For example, Freathy et al. (2009) show that the common polymorphismrs1051730 in the nicotinic acetylcholine receptor gene cluster (CHRNA5-CHRNA3-CHRNB4) is associated with heavier smoking in the first, but not third trimester. We will explore the effects of these different timings on our outcomes of interest.
We will use the variants CHRNA3 (rs1051730) and ADH1B (rs1229984) as instruments for smoking and alcohol consumption respectively. Finally, FTO (rs9939609) and MC4R (rs17782313) will be used as instruments for maternal BMI. Within the ALSPAC cohort, data have already been collected on a series of SNPs, with robust evidence that these SNPs affect these risky behaviours, which has been replicated in different independent samples (see e.g. Freathy et al., 2009; Zuccolo et al, 2009; Frayling et al., 2007; Loos et al., 2008). As such, we wish to use these in IVQR analyses, allowing us to specifically examine the distributional effects on these outcomes of interest.
We will also discuss the potential limitations of power within this setting. We are aware that (a) the minor allele rs1229984 is relatively rare in the ALSPAC population (MAF 2.4%), and that (b) the genetic effects on the intermediate trait are relatively small. Hence, depending on the variation required to shift the outcome of interest, this may lead to low power. The implications will differ depending on the outcome of interest, which is something we will discuss in the paper. For example, it may be easier to find effects on child birth weight compared to effects on their educational outcomes.
As the principle applicant already has access to the ALSPAC data on child and family background characteristics, we only wish to obtain (a) the more detailed measures of maternal smoking and drinking and (b) the genetic markers related to mother's BMI, smoking and drinking.
B1036 - Longitudinal modelling of adiposity its determinants and its health consequences across childhood and adolescence Fellowship - 07/09/2010
Background
Obesity is associated with increased risks of morbidity and mortality, and globally it represents a considerable burden for healthcare systems. The prevalence of obesity in children and adolescents has risen dramatically in recent decades across most western countries and several low-income countries.1 Although some recent data suggest prevalence may have stabilised in the USA, UK and Sweden,2-4 it is too soon to know whether this is a true abating of the epidemic and levels remain high in all of these countries. A life course perspective is essential in the study of obesity, as emphasised by the MRC obesity research strategy published in 2010. There are many unanswered questions in the field of life course influences on obesity, which it is important to address in order to further the understanding of the aetiology of obesity and therefore to inform the design of prevention and treatment initiatives. Although we know that overweight tends to track across the life course5 and that childhood overweight and obesity is associated with adverse health outcomes in adulthood6, few studies have had sufficiently detailed longitudinal data to model individual trajectories of adiposity across childhood and adolescence and therefore to examine in detail the nature of adiposity changes across childhood and adolescence and how these changes relate to the determinants and health consequences of obesity. There is evidence to suggest that obesity-related cardiovascular risk factors such as blood pressure and lipids track across the life course within individuals7, 8, and that the associations of these risk factors with atherosclerosis in adulthood are similar regardless of whether the risk factors are measured in childhood or in adulthood at the same time as assessment of atherosclerosis.9 Again, however, few studies have had sufficient data to model the development of these obesity-related cardiovascular risk factors across childhood and adolescence and to relate them to patterns of adiposity change. Longitudinal models of behavioural risk factors for obesity such as diet and physical activity are also relatively rare in the literature, and the interplay between multiple genetic, lifestyle, and socioeconomic risk factors for obesity and longitudinal trajectories of adiposity has not been well explored. It would be of great public health interest to use longitudinal models of changes in adiposity, obesity-related cardiovascular risk factors, and lifestyle risk factors for obesity to identify whether there are subgroups of the population who could be identified in early life as at risk of extreme obesity and associated cardiovascular risk in adolescence. Furthermore, little is known about why some obese individuals remain metabolically healthy; patterns of change in adiposity, cardiovascular risk factors and lifestyle factors could be explored in such individuals to explore the issue of 'protective phenotypes' in obesity - identified as a priority area in the MRC obesity research strategy. I propose to address these areas of life course epidemiology of obesity using longitudinal modelling of adiposity, its determinants and its health consequences across childhood and adolescence using data from prospective cohort studies.
Aim 1: To characterise patterns of change in adiposity and its associated cardiovascular risk factors across childhood and adolescence
Specific research questions:
1. How does adiposity change across childhood and adolescence, as measured by BMI (0-18 years), waist circumference (7-18 years) and directly determined total body fat mass (9-18 years)?
2. Do changes in BMI, waist circumference and directly determined fat mass from age 9 to 18 mirror each other or are there periods where change between these differs? For example does timing of change in waist differ from that in general adiposity (BMI or total fat mass) around the time of puberty?
3. How does blood pressure change from age 7-18 years?
4. How do lipids, insulin, and inflammatory markers change from age 9-18 years?
Aim 2: To identify lifestyle and socioeconomic determinants of adiposity changes across childhood and adolescence
Specific research questions:
1. What are the joint effects of trajectories of physical activity (measured using accelerometers) and diet (measured using food frequency questionnaires and diet diaries) from ages 7-13 years on changes in adiposity from ages 7-18 years?
2. What are the associations of changes in socioeconomic position (e.g. differences between grandparental, parental and own education) on changes in adiposity from birth to 18 years?
3. Are there socioeconomic differences in trajectories of physical activity (measured by accelerometer) and diet (measured by diet diaries and food frequency questionnaires) across childhood and adolescence? Are these inequalities of consistent magnitude across childhood and adolescence and to what extent do they mediate socioeconomic inequalities in adiposity trajectories?
Aim 3: To identify how changes in adiposity relate to changes in cardiovascular risk factors in childhood and adolescence
1. How do changes in BMI (birth to 18), waist circumference (ages 9-18), and fat mass (ages 9-18) relate to changes in blood pressure (ages 7-18), insulin, lipids and inflammatory markers (ages 9-18) in childhood and adolescence?
2. Do increases in markers of adiposity precede increases in cardiovascular risk factors (blood pressure, insulin, lipids and inflammatory markers) and if so what is the lag period between adiposity change and risk factor change for each risk factor?
Aim 4: To develop predictive models of extreme obesity and of metabolic resilience to obesity
Specific research questions:
1. Are there specific subgroups of children/adolescents with particularly adverse changes in adiposity, cardiovascular risk factors and lifestyle factors who could be identified in early life as at risk of extreme obesity?
2. Are there some overweight adolescents who remain more metabolically healthy than other overweight adolescents? If so, what characterises these individuals with respect to individual trajectories of adiposity, cardiovascular risk factors and lifestyle factors across childhood and adolescence?
Statistical analysis:
As part of my current role, I have modelled trajectories of height, weight, and ponderal index/BMI from birth to ten years using random effects linear spline models (in application put references to your papers here). In order to extend the models for BMI to age 18 and to model trajectories of waist circumference from ages 7-18 and DXA-determined total body fat mass from ages 9-18, a different modelling strategy will be required owing to individual variation in the timing of puberty onset. A number of potential modelling strategies could be used for these changes, with each having different strengths and limitations. I therefore propose to model BMI, waist circumference and fat mass from their earliest age of measurement (birth, 7 and 9 respectively) through to age 18 in several different ways. I will compare how trajectories derived by these different approaches associate with potential determinants and consequences. Trajectories of blood pressure, lipids, insulin and inflammatory markers will be modelled using random effects linear spline models in a similar way to the models I have already constructed for height, weight and adiposity to age 10. Trajectories of physical activity are currently being modelled by a colleague (Alex Griffiths); these will be available for use by the start of this fellowship. I will analyse longitudinal diet patterns in multilevel models using repeat scores from principal component analysis of dietary patterns using data from diet diaries and food frequency questionnaires. Associations between adiposity and cardiovascular risk factor trajectories and their determinants (physical activity, diet, etc) will be modelled using multivariate models. The relationship between adiposity trajectories and cardiovascular risk factor trajectories will be modelled in a joint multivariate model. Predictive models for 1) extreme obesity in adolescence and 2) metabolic resilience to obesity will be developed and internally validated. Accuracy of prediction will be assessed by discrimination using area under the receiver operator curves (AUROC) and calibration by comparing observed to predicted risk by tenths of the predicted risk score.
Relation to existing ALSPAC approved projects:
Aims 1-3 are already have approval under the ALSPAC exec approval for the health strand of the ESRC large grant (strand leader: Debbie Lawlor). The aims of that strand were somewhat ambitious and it will not be possible to complete all of them during the period of the grant and hence this application sensibly extends the aspect of that grant related to child obesity. The additional analyses proposed for this fellowship build on work that I have already completed in my position as RA on the ESRC large grant. Aim 1 is also already partly covered by the MRC life course grant for which Kate Tilling is PI and which already has ALSPAC exec approval. Only Aim 4 is completely novel and not covered by the existing approval.
Plan for completing work:
1. LH, DL and KT will agree specific objectives and analysis methods
2. LH will put together datasets and conduct analyses (LH already has approved access to ALSPAC built datasets)
3. LH will draft initial manuscripts
4. LH, DL, KT and other co-authors will be involved in study design, and critical evaluation of manuscripts as appropriate for each paper
All derived variables from the trajectories work will be returned to the exec. with appropriate documentation so that they can be widely used by others (as we have already done with the growth data from birth to 10).
B1035 - EAGLE Motor Development GWAS meta-analysis - 07/09/2010
1. Background
The aim of this study is to perform a meta-analysis of GWAS results on the date of achievement of independent locomotion during infancy. Understanding the processes underlying the timing and achievement of developmental motor milestones is critical, since they are considered to be important indicators of the child's neurological, psychomotor, and socio-emotional integrity and subsequent development. Finnish and UK longitudinal cohort studies have highlighted the importance of infant motor development by demonstrating that failure to achieve this developmental milestone at the expected age is a strong predictor of neuropsychiatric conditions, reduced physical performance and fitness over the lifecourse1,2,3,4,5,6.
Due to the relatively brief exposure of the infant to environmental influences (compared to later periods of life) most studies focusing on early developmental milestones have estimated high heritabilities. For example, although motor development throughout childhood and adolescence is expected to be a process that develops through interactions of the individual with the environment7, twin studies on infants have estimated that over 90% of the variation in motor milestone achievement is influenced by genetic factors8. This was also recently highlighted through the first GWAS study on an aspect of infant development, tooth development, a study which demonstrated the potential to identify a large number of genetic variants through GWAS meta-analysis of relatively small sample sizes9.
The study will focus on the complete, normal, distribution of the date of achievement of independent locomotion (analysis I), but also examine the delayed achievement of independent locomotion (analysis II). This is because there is support for a larger genetic contribution to delayed as compared to non-delayed development, as heritability estimates, e.g. for early language development, increase from 21% to 48% when shifting the focus from normal to delayed development10.
2. Traits of interest
For analysis I: Walking without support/unaided (age in months)
For analysis II: Walking without support unaided at 18 months (can walk - yes/no)
3. Participating studies and sample numbers
Table 1. Individuals with GWAS and phenotype data (as of 11-07-2010)
Study
Analysis I
Analysis II
NFBC1966
4000
4000
RAINE
1500
1500
MoBa
630
630
NCDS1958
-
4500
ALSPAC
-
3000
Generation R
-
2650
Tweeligenregister
?
?
Total projected sample
6130
16280
https://spreadsheets.google.com/ccc?key=0AgoCkeBwIrX4dDVjWFYtR1NQM1pfUVBrQXAyYlBBZEE&hl=nl#gid=8
2. Phenotypes / variables necessary from ALSPAC dataset
i) At 18 months: F4(d) -greater than child ' can walk alone for at least 5 steps' (Motor milestones,Denver)
ii) For exclusion:
- intellectual disability (which infants are intellectually disabled) (if availlable)
- cerebral palsy (which infants suffered from cerebral palsy) (if availlable)
iii) gestational age (covariate to be adjusted)
B1033 - Early feeding of cows milk in relation to growth and bone development in ALSPAC - 07/09/2010
Early feeding of cows' milk in relation to growth and bone development in ALSPAC
The current guidelines on infant feeding state that cows'milk should not be given as the main milk drink to infants before the age of 12 months. Despite this in the latest infant feeding survey 6% of 9 month olds were being fed in this way. In ALSPAC when the CIF children were 8 months old diet records were provided for 1178 infants - 13% of these were having cows' milk as their only source of milk at this age and half of these were having more than 600 mg per day. In this group, in particular, the diet was much higher in energy, protein and fat than breast fed infants and there was evidence that these children grew faster up to 61 months of age. This group also had a very high calcium intake compared to the breast fed group and this might lead to greater bone density in later life. However we showed that levels of anaemia were high in the cows' milk fed infants and their intakes of many other nutrients were low. Therefore it is important to investigate further the long-term effects of early introduction of cows' milk in a longitudinal cohort.
Planned analysis
Using questionnaire data at 15 months we will devise feeding groups in relation to the early introduction of cows' milk. The groups identified will be mutually exclusive
1. Breast feeding at 6 month or beyond (reference category)
2. Cows' milk introduced by 6 months (breast feeding stopped before 6 months)
3. Cows' milk introduced between 7-12 months (breast feeding stopped before 6 months)
4. Cows' milk introduced at or after 12 months (breast feeding stopped before 6 months)
We will use CIF diet data at 8 months to verify that these questionnaire identified groups are likely to be associated with meaningful dietary differences in the way infants were fed.
We will investigate the following outcomes in relation to these milk groups:
Bone density at 9 and 11 years from DXA measurements made in focus clinics
Anthropometric measurements at 7, 9 and 11 years
Confounding variables such a sex of the child, parity, maternal smoking in pregnancy and education level will be included
B1028 - Predictors of Persistence and Psychological Impact of childhood Eczema POPPIE - 06/09/2010
AIM: To identify factors that predict persistent atopic eczema and/or negative psychological outcomes in children with atopic eczema.
BACKGROUND: Atopic eczema (AE, otherwise known as atopic dermatitis) is one of the commonest disorders in childhood. The disease is characterized by generalized dry skin, pruritis, and typically flexural erythema. Symptoms usually start before two years of age, and in the UK, most children will be diagnosed and treated by their general practitioner. Whilst the majority of children will have mild symptoms that improve as they get older, a significant proportion of those affected will have more severe and persistent eczema.
Our knowledge about the natural history of AE and factors that might predict persistence beyond the pre-school years is limited by the methodological imperfections of research conducted to date. Most studies have either been small, retrospective case series; suffered from a short duration of follow-up; or recruited participants from hospital in or out-patients.1 As a consequence the findings may be unrepresentative of children in the community and subject to the many biases that are associated with a retrospective study design. The only two significant cohort studies are by Williams and Strachan2 and Illi.3 Early onset of disease, severity, atopic sensitization and a strong atopic family history were associated with a worse prognosis.
As well as the physical burden of the disease, a series of studies have suggested AE can have a significant psychological impact as well. However, nearly all of these studies have serious limitations of one form or another, related to size of the study, the population from which their sample was drawn (mainly hospital out-patients), the design of the study (most are cross-sectional) and/or the outcome examined (the majority examine the psychological effect of AE on the child only). Only one recently published study by Schmitt et al4 followed-up a sizeable cohort of children for a reasonable length of time. It reported that children with infant-onset eczema were more likely to have emotional disturbance, but data on psychological health were collected just at one time-point (age 10 years) and only on the affected child.
This project will build on earlier research that has examined children in ALSPAC who have AE5-8 and address many of the limitations of previous studies that have examined prognostic factors and the psychological consequences of AE. Specifically, participants in ALSPAC were recruited from the community rather than specialist care settings; relevant data has been collected prospectively from several sources (parent, school and physical examination) with good rates of follow-up; data are available into adolescence and beyond; and clear definitions of AE can be applied using previously described criteria.5,6 For the first time will we link data available on both child and parental mental health to persistent eczema, to assess the impact of AE on both the child and the family.
In addition to answering the research questions posed, one of the outputs from this project will be to produce a map all of the questions relevant to AE. Serial questionnaires are the main source of data in ALSPAC and a detailed manual identifies which psychological instruments and have been used when. Unfortunately there in no comparable document that describes the all the different questions relating to AE. By comprehensively documenting for the first time the method of collection, timing and nature of data on AE, a valuable resource will be made available to future researchers wanting to interrogate ALSPAC for studies of AE.
Finally, we will extend the analysis performed by Wadonda-Kabondo et al5 on the prevalence and incidence of AE between to include children up to 128 months old.
PLAN OF INVESTIGATION:
We will examine data on data on parents and children both antenatal and up to 12 years of age. We have chosen this upper age limit because it is a natural watershed between childhood and early adolescence; it fits with age range of the 2007 NICE guidance; and because it represents a significant amount of data that can be explored, with potential for important findings, within the proposed research timeframe and budget. Also, previous work suggests that after the age of 11 years, the proportion of children whose eczema persists is relatively constant.2 However, once the methods have been established, dependent of course upon the findings of this study and subject to application for further funding, the approach applied to the 0-12 year age group could of course be extended to look at the 13-18 year age group and upwards.
Definition of AE: We will adopt the case of definition of AE used by Wadonda-Kabondo et al:5,6 children with a reported rash on at least two separate occasions. This was found to give estimates of the prevalence and cumulative incidence of AE similar to those from other studies.9,10 We will further sub-divide children with AE into three groups on the basis of whether AE is reported before or after 2 years of age: infant onset clearing (present before age 2 years but not after), infant onset persistent (present before age 2 years and after), or late onset (present after age 2 years but not before). Similar categories have been adopted by Illi et al3 and Schmitt et al4 and permits analysis of data and reporting of these sub-groups in a way that is reflects the patterns of AE clinicians commonly see.
Psychological measures: The psychological health of the child will be assessed using the Strength and Difficulties Questionnaire (SDQ),11 which was completed by parents at 4, 7, 8 and 11 years and the child's school at 8 and 10 years. SDQ scores above the 90th centile indicate problem behaviours and are associated with a substantially increased risk of psychiatric disorder.11 Parental psychological (depression and anxiety) health will be assessed using the Edinburgh Post-Natal Depression Score (EPDS)12 and the anxiety items from Crown-Crisp Experiential Index (CCEI).13 Mothers and father complete EPDS and CCEI approximately annually up to 11 years of age.
Analysis: 1) All of the questionnaires will be reviewed for questions pertaining AE and previously described risk factors for onset and persistence of AE. Undertaking this preliminary work will be important to both this study but also in establishing a map of atopy and skin questions in the ALSPAC dataset that will be of use in future ALSPAC-based research of AE. 2) Based on parental report, we will calculate and report the incidence and prevalence of AE at different time-points. We will also examine the reliability of the reporting of eczema. First, by comparing responses to items in earlier questionnaires that ask about skin rashes with later data that asks about "ever diagnosis" of eczema. Second, by comparing reported prevalence of AE with that recorded at clinic examination. In doing this, we will extend the analysis performed by Wadonda-Kabondo and colleagues.5 3) Using the definitions of AE given above, using simple descriptive statistics we will describe the different groups of children with and without AE. Next will we examine the data for associations between AE and characteristics which may be risk factors for AE, including birth weight, gestational age, breast-feeding, early age of onset, flexural site at disease onset, female sex, allergy (peanut, egg, fish or dust mite), hay fever or asthma, social class, family size, and family history of atopy. 4) The prevalence of child and parental psychological distress at different time-points will be reported. We will compare psychological measures for children with and without AE, adjusting for known confounders; and compare psychological measures for parents (mother and partner if available) of children with and without AE, adjusting for known confounders. These analyses will be repeated for the different sub-types of AE described. Sensitivity analyses will be conducted to examine the influence of missing data on the findings
DISSEMINATION OF FINDINGS:
A final report will be sent to the British Skin Foundation and other interested professional and patient representative bodies, including the National Institute of Health and Clinical Excellence (NICE, the guideline "Atopic eczema in children" is due to be revised in 2011), relevant Royal Colleges, the University of Nottingham Centre for Evidence Based Medicine, British Association of Dermatologists, the Primary Care Dermatology Society, and the National Eczema Society. The findings will also be published in peer-reviewed scientific journals.
ABOUT THE STUDY TEAM:
Matthew Ridd is an GP and lecturer in primary care, with an interest in promoting research to understand the nature and best management of common dermatological problems seen in primary care. He will be responsible for supervising an RA who will lead on the project; monitoring budgets and progress; and producing the final report.
Sarah Purdy is a consultant senior lecturer and practicing GP. She has conducted primary research on the epidemiology and management of acne in primary care and secondary research on the management of acne and eczema in primary care. She is a clinical epidemiologist and has extensive experience of working with large datasets.
Sarah Sullivan has experience of using the ALSPAC data set, especially in relation to measures of mental health. Her PhD is using the ALSPAC cohort to examine Psychosis-like Symptoms (PLIKS) and Social Functioning.
GD is a consultant dermatologist who co-supervised Dr Wadonda-Kabondo's PhD, from which two of the most significant studies of AE using the ALSPAC dataset have been published.
B266 - Long-term effects of exposure to oxytocin - 06/09/2010
The research question concerns the use of oxytocin and the way in which that might affect the mother's interaction with her child. This requires assessment of details available in medical records [the obstetric record extraction only obtained a yes/no answer to this, but duration and dose are required].