The Definition of Childhood Obesity

The Definition of Childhood Obesity

 Childhood adiposity can be measured in numerous ways although body mass index (BMI) remains the most commonly used. Whilst BMI is a relatively simple tool with which to assess body mass, it is a relatively poor predictor of actual body composition in both adults and children. Due to its ease of determination however, along with a good correlation with body fat, it has remained the accepted method to define obesity in children based on current expert opinion [5]. Another important measure is waist circumference which has been validated as a surrogate marker of visceral adiposity in children [6]. In adults, a BMI 25 and 30 corresponds to ‘overweight’, whereas a BMI 30 identifies those with obesity. These cut-off points correspond to an increased risk of cardiovascular disease and diabetes in adults. In children, however, BMI changes with normal longitudinal growth, as shown in figure 1. Therefore, it is inappropriate to simply express raw BMI in children, without adjusting for age and sex. Instead, BMI standard deviation scores or z-scores (BMI SDS – representing increases or decreases around the 50th centile for age and sex) are used to determine which children are relatively ‘overweight’ or ‘obese’. For national statistics, BMI levels of 95th, 97th, or 98th percentile have been used to identify the ‘fattest’ children within different populations. These limits have high specificity and moderate sensitivity, and allow temporal changes within countries to be assessed. For across-country comparisons, international cut-off points must be used and these have been based on the extrapolation of adult cut-off points back into childhood [7]. The international cut-off points tend to greatly underestimate obesity prevalence when used for determining prevalence rates within a specific country [8]. However, BMI SDS may not be the best tool with which to assess longitudinal changes in adiposity in children enrolled into weight management programs, as the within-child variability over time depends upon the child’s levels of adiposity. Under these circumstances, age-adjusted BMI (calculated by subtracting the sexand age-specific median BMI) may be a better tool [9]. However, the normative data necessary to make these calculations are currently unavailable so most workers continue to use BMI SDS when reporting their results.