Resemblance between child & parental eating, physical activity & obesity patterns

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): The overall goal of this 2-year secondary data analysis R03 project is to study the resemblances between American children's and their parents' eating, physical activity (PA) patterns, and body weight status (obesity), and the influence of individual-, household-, and environmental factors on the resemblances. A good understanding of the familial resemblances and discrepancies, related between-population differences, and the determinants of the resemblances based on nationally representative data will help understand the etiology of childhood obesity, determinants of children's eating and PA behaviors and the related health disparities, provide useful insights for the development of population-based interventions to help achieve the Health People 2010 national goals. In the proposed study, related nationally representative data collected in three surveys, including longitudinal data, to be linked with additional contextual factor measures from other data sources using geocode will be used. The first three aims are our primary objectives, the others are secondary objectives: Aim 1: Examine the child-parent resemblance in dietary intakes. We will focus on a number of indicators of dietary quality including intakes of food groups, energy and nutrients, and an overall dietary quality score. Aim 2: Study the differences in the child-parent resemblance in intakes by individual-, household-, and environmental factors. This will help increase our understanding of the influence of the latter on the resemblance. A set of analyses will be conducted to address several questions. Aim 3: Study the child-parent resemblance in BMI and weight status (obesity) and the influence of the multi-level factors on the resemblance. We will also test if the resemblance has changed over time and will test the resemblance in BMI change based on repeated measures. Aim 4: Study the child-parent resemblance in PA and sedentary behaviors and the influence of multi-level factors. Aim 5: Assess the influence of measurement error on findings of the resemblance in dietary intakes, e.g., by comparing the measurement error corrected- ('calibrated') versus uncorrected results. A set of sophisticated statistical approaches such as multivariate linear and logistic regression analysis, multilevel models, and structural equation models will be used to achieve our goals, in particular, to examine the influence of the predictors. As the first such study based on national data and innovative analysis approaches, our findings will help fill a major gap in the literature and have many important public health implications.