Childhood obesity is a major public health problem facing our country today, affecting the health of over 17 million youth. However, a major challenge to effectively addressing childhood obesity is collecting accurate measures of body mass index (BMI), which is the most suitable and feasible measure of obesity at a population level. Professionally measured heights and weights are the gold standard for measuring BMI. In large population-based studies, professionally measured BMI may not be feasible;self-reported estimated heights and weights, often reported by a parent, become the default. Classifying children as overweight/obese using parent estimates of child weight status is prone to extremely high rates of misclassification. Given the time and cost intensiveness of professionally measured heights and weights, the tendency for misclassification with parent-estimated heights and weights, and the growing need to accurately assess weight status at lower costs, alternative strategies need to be identified. A novel alternative approach would be to request that parents measure their children's heights and weights using an easy-to-use protocol, but to-date, no studies have tested the accuracy of such an approach. Thus, the aim of this study is to compare the accuracy of classifying the weight status of children using parent-measured heights and weights over parent estimated heights and weights among 300 urban youth from a population-based sample. We will assess the extent of the misclassification with each approach (parent-measured vs. parent-estimated) against the gold standard of professionally measured heights and weights. We expect to observe significant improvements in classifying children as normal versus overweight/obese using parent-measured heights and weights over parent estimates. In addition, we will explore potential biases in parent measured heights and weights by parent and children characteristics (e.g., income, education, gender, race). This R03 study will capitalize on a NICHD-funded R01 grant that is studying environmental determinants of obesity among 1200 low-income youth in New Jersey. We will draw a subsample of children (n=300) from the parent R01 study and augment data collection efforts by providing participating parents with pre-tested tools and instructions to measure their children's heights and weights. We will also provide resources for professional measurements of heights and weights on the same 300 children. Data on children's heights and weights, using three different approaches (parent-estimates, parent-measured, and professionally-measured), will be collected within a 4-week period. Findings from this study will provide important results about the use of a low-cost method for collecting accurate measurements of heights and weights, increasing the sensitivity of classification of weight status in large community-based sample. Findings will also provide critical information on cost relative to error, which will help researchers in the field select the most accurate and cost effective approaches for collecting BMI data.
Innovative, accurate, and cost effective approaches to collecting data on heights and weight are needed in order to efficiently assess weigh status among children, specifically when professionally measured data collection are not feasible or affordable. This study will compare the accuracy of child weight status classification using parent-measured heights and weights and parent-estimated heights and weights against the gold standard of professional measurements. Findings will provide critical information on cost relative to error, which will help researchers select the most accurate and cost effective approaches for collecting BMI data.
