Entity

Time filter

Source Type


Hughes A.R.,University of Stirling | Sherriff A.,University of Glasgow | Lawlor D.A.,Center for Causal Analyses in Translational Epidemiology | Ness A.R.,University of Bristol | Reilly J.J.,University of Glasgow
Pediatrics | Year: 2011

OBJECTIVES: To test the hypothesis that most excess weight gain occurs by school entry in a large sample of English children, and to determine when the greatest gain in excess weight occurred between birth and 15 years. METHODS: Longitudinal data were collected annually from birth to 15 years in 625 children. Weight and BMI at each time point were expressed relative to UK 1990 growth reference as z scores. Excess weight gain was calculated as the group increase in weight and BMI z scores between specific time periods. RESULTS: Weight z score did not increase from birth to 5 years (mean difference: 0.04 [95% confidence interval (CI): -0.03- 0.12] P=.30) but increased from 5 to 9 years (mean difference: 0.19 [95% CI: 0.14-0.23] P < .001). BMI z score increased from 7 to 9 years (mean difference: 0.22 [95% CI: 0.18-0.26] P<.001), with no evidence of a large increase before 7 years and after 9 years. CONCLUSIONS: Our results do not support the hypothesis that most excess weight gain occurs in early childhood in contemporary English children. Excess weight gain was substantial in mid-childhood, with more gradual increases in early childhood and adolescence, which indicates that interventions to prevent excess weight should focus on school-aged children and adolescents as well as the preschool years. Copyright © 2011 by the American Academy of Pediatrics. Source


Hughes A.R.,University of Stirling | Sherriff A.,University of Glasgow | Lawlor D.A.,Center for Causal Analyses in Translational Epidemiology | Ness A.R.,University of Bristol | Reilly J.J.,University of Strathclyde
Preventive Medicine | Year: 2011

Background and Aims: Timing of obesity development during childhood and adolescence is unclear, hindering preventive strategies. The primary aim of the present study was to quantify the incidence of overweight and obesity throughout childhood and adolescence in a large contemporary cohort of English children (the Avon Longitudinal Study of Parents and Children, ALSPAC; children born 1991-1992). A secondary aim was to examine the persistence of overweight and obesity. Methods: Longitudinal data on weight and height were collected annually from age 7-15. years in the entire ALSPAC cohort (n = 4283), and from 3 to 15. years in a randomly selected subsample of the cohort (n = 549; 'Children in Focus' CiF). Incidence of overweight and obesity (BMI (Body mass index) at or above the 85th and 95th centiles relative to UK reference data) was calculated. Risk ratios (RR) for overweight and obesity at 15. years based on weight status at 3, 7, and 11. years were also calculated. Results: In the entire cohort, four-year incidence of obesity was higher between ages 7 and 11. years than between 11 and 15. years (5.0% vs 1.4% respectively). In the CiF sub-sample, four-year incidence of obesity was also highest during mid-childhood (age 7-11. years, 6.7%), slightly lower during early childhood (3-7. years, 5.1%) and lowest during adolescence (11-15. years 1.6%). Overweight and obesity at all ages had a strong tendency to persist to age 15. years as indicated by risk ratios (95% CI (Confidence interval)) for overweight and obesity at 15. years from overweight and obesity (relative to healthy weight status) at 3. years (2.4, 1.8-3.1), 7. years (4.6, 3.6-5.8), and 11. years (9.3, 6.5-13.2). Conclusion: Mid-late childhood (around age 7-11. years) may merit greater attention in future obesity prevention interventions. © 2011 Elsevier Inc. Source


Gage S.H.,University of Bristol | Gage S.H.,Center for Causal Analyses in Translational Epidemiology | Zammit S.,University of Bristol | Zammit S.,University of Cardiff | Hickman M.,University of Bristol
F1000 Medicine Reports | Year: 2013

Schizophrenia is a debilitating but poorly understood condition with very few known modifiable risk factors. Cannabis use can acutely induce psychotic experiences, but its causal relationship to schizophrenia is less well understood.Longitudinal cohort studies suggest that the association between cannabis and psychotic outcomes isnot due to chance or reverse causation. However, the association could be due to bias or residualconfounding. Methods that can test alternative explanations in greater depth are required. This is especially important as ecological studies have found little association between the increase in cannabis use over recent decades and incidence of psychotic disorders; public health models suggest that cannabis use may need to be treated and prevented in many thousands of users in order to prevent one case of schizophrenia. We believe that, while such uncertainty exists, there is a scientific duty to continue to investigate the role of cannabis in the aetiology of schizophrenia and that the policy case for considering cannabis exposure as a critical target for preventing schizophrenia is yet to be made. However, due to other evidence of the harms of cannabis use, this should not affect the public health message that cannabis can be harmful and that cannabis dependence should be prevented. © 2013 Faculty of 1000 Ltd. Source


Seoane J.A.,Center for Causal Analyses in Translational Epidemiology | Day I.N.M.,Center for Causal Analyses in Translational Epidemiology | Gaunt T.R.,Center for Causal Analyses in Translational Epidemiology | Gaunt T.R.,University of Bristol | Campbell C.,University of Bristol
Bioinformatics | Year: 2014

Motivation: Within medical research there is an increasing trend toward deriving multiple types of data from the same individual. The most effective prognostic prediction methods should use all available data, as this maximizes the amount of information used. In this article, we consider a variety of learning strategies to boost prediction performance based on the use of all available data.Implementation: We consider data integration via the use of multiple kernel learning supervised learning methods. We propose a scheme in which feature selection by statistical score is performed separately per data type and by pathway membership. We further consider the introduction of a confidence measure for the class assignment, both to remove some ambiguously labeled datapoints from the training data and to implement a cautious classifier that only makes predictions when the associated confidence is high.Results: We use the METABRIC dataset for breast cancer, with prediction of survival at 2000 days from diagnosis. Predictive accuracy is improved by using kernels that exclusively use those genes, as features, which are known members of particular pathways. We show that yet further improvements can be made by using a range of additional kernels based on clinical covariates such as Estrogen Receptor (ER) status. Using this range of measures to improve prediction performance, we show that the test accuracy on new instances is nearly 80%, though predictions are only made on 69.2% of the patient cohort. © 2013 The Author 2013. Published by Oxford University Press. Source


Paternoster L.,Center for Causal Analyses in Translational Epidemiology | Howe L.D.,Center for Causal Analyses in Translational Epidemiology | Tilling K.,University of Bristol | Weedon M.N.,University of Exeter | And 9 more authors.
Human Molecular Genetics | Year: 2011

Previous studies identified 180 single nucleotide polymorphisms (SNPs) associated with adult height, explaining ~10% of the variance. The age at which these begin to affect growth is unclear. We modelled the effect of these SNPs on birth length and childhood growth. A total of 7768 participants in the Avon Longitudinal Study of Parents and Children had data available. Individual growth trajectories from 0 to 10 years were estimated using mixed-effects linear spline models and differences in trajectories by individual SNPs and allelic score were determined. The allelic score was associated with birth length (0.026 cm increase per 'tall' allele, SE = 0.003, P = 1 × 10 -15, equivalent to 0.017 SD). There was little evidence of association between the allelic score and early infancy growth (0-3 months), but there was evidence of association between the allelic score and later growth. This association became stronger with each consecutive growth period, per 'tall' allele per month effects were 0.015 SD (3 months-1 year, SE 5 0.004), 0.023 SD (1-3 years, SE = 0.003) and 0.028 SD (3-10 years, SE = 0.003). By age 10, the mean height difference between individuals with ≤170 versus ≥191 'tall' alleles (the top and bottom 10%) was 4.7 cm (0.8 SD), explaining ~5% of the variance. There was evidence of associations with specific growth periods for some SNPs (rs3791675, EFEMP1 and rs6569648, L3MBTL3) and supportive evidence for previously reported age-dependent effects of HHIP and SOCS2 SNPs. SNPs associated with adult height influence birth length and have an increasing effect on growth from late infancy through to late childhood. By age 10, they explain half the height variance (~5%) of that explained in adults (~10%). © The Author 2011. Published by Oxford University Press. Source

Discover hidden collaborations