Predictive performance of the waist-to-height ratio and the conicity index for diagnosing excess body fat in children

Desempenho preditivo da razão cintura-estatura e do índice de conicidade para diagnóstico de excesso de gordura corporal em crianças

ABSTRACT

Objective:  To evaluate the performance of waist-to-height ratio (WHtR) and conicity index (CIndex) as predictors of body fat percentage (BF) in children, as well as to determine the most appropriate cut-off points for this population.

Methods:  This school-based study assessed 314 children aged five to nine years. Body composition was evaluated using bioelectrical impedance analysis (BIA), and children were classified as having excess or normal BF according to sex. WHtR and CIndex were calculated using their respective formulas. To assess the predictive power of CIndex and WHtR, the area under the curve (AUC) of Receiver Operating Characteristic was analyzed, using excess BF by BIA as the gold standard.

Results:  WHtR was considered a good predictor of excess BF in both sexes, with AUC values greater than 0.8. CIndex was classified as a poor predictor (AUC<0.7 in both sexes). WHtR showed a significantly higher AUC than CIndex (p<0.001), according to the Wald test. Given the higher accuracy of WHtR, the optimal cut-off points to identify excess BF in children were >0.47 for boys and >0.45 for girls.

Conclusions:  WHtR demonstrated good performance in predicting excess BF in the children evaluated and may be included in child health surveillance as an indicator of alterations in body composition.

Keywords:
Body composition; Pediatrics; Overweight; Obesity; Students; Waist-height ratio