BODYFAT: a new calculator to determine the risk of being overweight validated in Spanish children between 11 and 17 years of age
María Victoria Martín-Miguel,María Victoria Delgado-Martín,Carolina Barreiro-Arceiz,Ana Goicoechea-Castaño,Sara Rodríguez-Pastoriza,Clara González-Formoso,Martín Fernández-Pérez,Clara García-Cendón,Javier Roca,Ana Clavería
DOI: https://doi.org/10.1007/s00431-024-05596-2
2024-06-20
European Journal of Pediatrics
Abstract:The assessment of body fat of children in primary care requires consideration of the dynamic changes in height, weight, lean mass, and fat mass during childhood growth. To achieve this, we aim to develop a predictive equation based on anthropometric values, with optimal diagnostic utility. This is a cross-sectional observational study, involving schoolgoers aged 11–17 years in the Vigo metropolitan area. Out of 10,747 individuals, 577 were randomly recruited. Variables: age, sex, ethnicity/country of origin, weight, height, 8 skinfolds, 3 diameters, 7 perimeters, and 85% percentile of body fat mass as the gold standard. Generalized additive regression was selected by cross-validation and compared using receiver operating characteristic curves (ROC curves). Sensitivity, specificity, positive and negative predictive values, true positive and true negative values, false positive and false negative values, accuracy, and positive and negative likelihood ratios were calculated. Two models were identified. The optimal model includes sex, weight, height, leg perimeter, and arm perimeter, with sensitivity of 0.93 (0.83–1.00), specificity of 0.91 (0.83–0.96), accuracy of 0.91 (0.84–0.96), and area under the curve (AUC) of 0.957 (0.928–0.986). The second model includes sex, age, and body mass index, with sensitivity of 0.93 (0.81–1.00), specificity of 0.90 (0.80–0.97), accuracy of 0.90 (0.82–0.96), and an AUC of 0.944 (0.903–0.984).
pediatrics