17th Annual ECSS-Congress, Bruges 2012

Abstract details

Abstract-ID: 991
Session: [MO-PM11] Physical Activity in Children 1
Lecture room: E108
Date & time: 02.07.2014 / 13:00 - 14:00
Title of the paper: A COMPARISON OF CHILD OBESITY PREVALENCE CHANGES USING INTERNATIONAL AND UNITED KINGDOM NATIONAL GROWTH REFERENCES FOLLOWING A 10-WEEK INTERVENTION
Authors: Brown, E.1, Cavana, J.2, Wyatt, F.3, Buchan, D.1, Baker, J.1, Kilgore, L.1
Institution: University of the West of Scotland
Department:
Country: United Kingdom
Abstract text Introduction BMI standard deviation scores (BMI-SDS) are commonly used in children, but a number of issues regarding statistical and epidemiological methods toward this approach of defining child obesity exist. The aim of this study was to compare prevalence changes from a 10-week child healthy weight intervention, Fit for School (FFS), based on International Obesity Task Force, UK1990 clinical, and UK1990 population surveillance child weight scales. Methods In order to compare the magnitude of the relationships between weight categories and across scales, the overall available differentiation and logarithmic magnitudes for prevalence changes were determined. To further examine the relationship of scales, Pearson correlations between scoring scales rates of change (ROC) were also calculated. Results Using the UK1990 population surveillance scale, a 7% decrease in the severely obese group, 1% decrease in the obese group, 0% change in the overweight group, and increases of 7% and 29% occurred in the underweight and severely underweight categories, respectively. When using the clinical UK1990ís definitions, a 1% decrease in severely obese, 10% decrease in obese, 3% decrease in overweight occurred, but underweight increased by 32 percent. The IOTF scale suggested a 2% and 1% decrease in the obese and overweight categories, respectively, and a 1% increase in the thinness grade 1 category with no change in the lower two thinness categories. Significant correlations of ROC were found between each of the scoring scales. There was a correlation of 0.490 between the UK1990 clinical and UK1990 population surveillance (p < 0.01), 0.277 between IOTF and UK1990 clinical (p < 0.01), and 0.295 between IOTF and UK1990 population surveillance. The use of 3 appropriately deemed BMI scales of child obesity for British children is problematic using the same data set yielding very different results in prevalence changes following an intervention. The results highlight the practical implications of using BMI as an indicator of child obesity. Using an accurate, easy to implement, and comparable measure of child obesity is paramount for decision making regarding government policy and funding. The aim of classifying child obesity should be to determine what degree of overfatness is associated with adverse health outcomes, not how a childís height and weight compares with peers. This may be accomplished through use of a criterion referenced evaluation such as waist-to-height ratio rather than BMI-SDS, a norm referenced evaluation. Contact Elise.brown@uws.ac.uk
Topic: Health and Fitness
Keyword I: BMI-SDS
Keyword II: Obesity
Keyword III: Health