[Obesity screening: updated criteria and their clinical and populational validity].

OBJECTIVES

The primary aim of this study was to compare the frequency of overweight and obesity in Spanish children and teenagers by using national and international standards for the body mass index (BMI). The second aim was to establish the association between criteria based on BMI and body composition. The third aim was to propose new standards of fat percentages in order to increase diagnostic accuracy in patients with excess weight.

MATERIAL AND METHODS

The sample was composed of 7,228 boys and girls aged between 6 and 20 years. Anthropometric measurements were taken (height, weight and subcutaneous skinfolds), and BMI and the percentage of fat were obtained using the methods of Durnin and Womersley, Siri and Slaughter et al. To classify obesity and overweight, the national standards of Hernández et al and the international standards published by Cole et al were used. The correlation between BMI and percentage of fat was established using Spearman's method and the variability fat percentage was estimated in relation to the BMI cut-off points proposed by the International Obesity Task Force (IOTF).

RESULTS AND CONCLUSIONS

Overweight and obesity were more frequent in boys than in girls and in the youngest individuals. According to national standards, between 9.6 % (6-12 years) and 9.5 % (13-20 years) of girls and between 10.62 % (13-20 years) and 12.50 % (6-12 years) of boys were overweight. According to international standards, between 4.68 % (13-20 years) and 6 % (6-12 years) of boys were obese and between 18.83 % (13-20 years) and 21.6 % (6-12 years) were overweight. The proportion of obesity in girls ranged from 2.81 % (13-20 years) to 5.9 % (6-12 years); the prevalence of overweight was 25.99 % in the group aged 6-12 years and was 14.55 % in that aged 13-20 years. The analysis carried out shows that the association between BMI and adiposity differs in normal individuals and in those with excess weight. The international standards of Cole et al tend to underestimate obesity and to overestimate overweight. Therefore, their usefulness is limited to comparative studies and their use cannot be recommended in clinical diagnosis, where it would be more effective to use of specific population standards, especially those for fat percentage.