Both studies will be carried out by the principal applicant, Dr Adrienne Hughes, during her 6 month sabbatical (August 2009 to January 2010) in collaboration with Professor John Reilly, Professor John McColl and Professor Andy Ness. Professor John McColl (Professor of Statistics, University of Glasgow) will provide advice on statistical analysis and interpretation. Adrienne's department has agreed to cover the costs incurred by ALSPAC as a result of this work.

Background

Early life events are thought to have important influences on later obesity and cardiometabolic health in children, suggesting that strategies to prevent childhood obesity should focus on young children. However, there is a lack of longitudinal studies using large cohorts of contemporary children. Thus, we intend to examine early life influences, such as timing of adiposity rebound and early weight/BMI gain on later obesity and metabolic health at age 15 years using a large cohort of contemporary children, the ALSPAC dataset.

STUDY 1

The aim of this study is to explore the associations between timing of adiposity rebound and later adiposity and cardiometabolic health at age 15 years.

Design and Methods

We intend to use the Children in Focus subsample (unless it is possible to calculate the timing of adiposity rebound for the entire cohort); the timing of adiposity rebound (AR) has previously been identified for the CiF subsample. In the Children in Focus subsample, 219 children had early AR (defined as before 61 months) and 563 children had later AR.

We require the adiposity rebound category (i.e. very early, early or late) for each child in the CiF subsample as well as length/height, weight and BMI from birth to 61 months.

Adiposity outcomes at 15 years will include: BMI (and external SD score), weight (and external SD score), waist, fat and fat free mass measured by DXA (adjusted and raw values), We also required height, age and gender.

Metabolic health outcomes at 15 years will include: blood pressure, fasting lipids, glucose and insulin.

We also require the following descriptors/potential confounding factors: maternal education and parental obesity.

Statistical Analysis

This will be informed by Professor John McColl, but is likely to involve the comparison of adiposity and metabolic health outcomes at age 15 years between children with early AR vs later AR.

Please note: The cardiometabolic health component (at age 15 years) of this study may have some overlap with Professor Debbie Lawlor and Professor Naveed Sattar's work, which has been discussed with Professor Andy Ness. Thus, we are willing to collaborate with Professor Debbie Lawlor and Professor Naveed Sattar if they and the ALSPAC Executive consider this to be appropriate.

STUDY 2

We intend to use the entire cohort of ALSPAC children as well as the Children in Focus subsample to address the following aims.

The first aim is to examine the persistence of overweight and obesity and progression from overweight to obesity over various time points (e.g. from preschool to age 15 years using Children in Focus, and age 7 years to age 15 years using entire cohort), and also to identify simple predictors of persistence and progression (e.g. parental obesity, socioeconomic status/maternal education).

The second aim is to describe the prevalence of obesity at specified ages (e.g. preschool, age 7, 11 and 15 years) and the incidence of obesity (i.e. development of new cases) over various time points (e.g. from preschool to 11 years using Children in Focus and age 7 to 11 years using entire cohort).

The final aim is to test the hypothesis (suggested in the Early Bird study) that most excess weight/BMI gain occurs by preschool/age 7 years. Early Bird (Gardner et al Pediatrics 2009) examined excess weight gain (change in weight SD score) between birth, 5 years of age and 9 years of age in small cohort of children. The study found that most excess weight was gained by 5 years of age, suggesting that obesity prevention strategies should focus on preschool children. However, this study involved only 233 children and followed children to only 9 years of age, thus we intend to examine how much weight/BMI gain has occurred by specified ages (e.g. preschool, age 7, 11 and 15 years years) using a much larger cohort of contemporary children.

Design and Methods

To address these aims, we intend to use data from the entire cohort of ALSPAC children as well as the Children in Focus subsample. For the entire cohort of ALSPAC children, we require the following data at birth and from age 7 years to age 15 years: weight (external z score), height/length, BMI (external z score), age, gender. We also require the following descriptors/potential confounders: parental obesity, maternal education, ethnicity.

For the Children in Focus subsample, we require the data described above at birth and from infancy and preschool to age 15 years.

Statistical Analysis

This will be informed by Professor John McColl.