Comparison of field methods to estimate fat mass in children

Background: Reliable field methods to measure fat mass (FM) in children may contribute to primary prevention of childhood obesity. Aim: The objective was to compare the accuracy of existing field methods (skinfold thickness (SF), leg-to-leg bioelectrical impedance analysis (BIA), anthropometrics for FM measurement in prepubertal European children. Subjects and methods: Reference FM was measured in 55 French children (30 boys, 25 girls; mean age 8.7 years) using a three-compartment model: body volume (BV) was assessed by air displacement plethysmography (ADP) and total body water (TBW) was assessed by deuterium dilution. Agreement between field methods and the reference method was assessed using Bland-Altman analyses. Since field methods for FM measurement are reported to be population-dependent, adjustment to the study population was performed using stepwise multiple linear regressions modelling. Results: Even after adjustment, field methods exhibited a high correlation (R(2) = 0.71-0.84) but a moderate agreement (+/- 3.32 to +/- 4.47 kg for fat mass) with the reference model. Methods based on BIA or SF performed slightly better than those based on anthropometry. Conclusions: Field methods for FM measurement may be recommended for epidemiological applications, but not for individual follow-up. New field equipment is required to improve accuracy of FM measurement in children and make individual follow-up possible.