THE INTEROBSERVER REPRODUCIBILITY OF BIOELECTRICAL-IMPEDANCE ANALYSISMEASUREMENTS IN INFANTS AND TODDLERS

Although nomograms to convert readings of resistance (R) and reactance (Xc) into estimates of body composition have begun to emerge for chil dren as well as adults, there has been reluctance to use bioelectrical impedance analysis (BIA) in very young children due to the irritabili ty of children, measurement instability, positioning of electrodes, an d, of course, reproducibility of measurements. The precision of comple tely independent measurements of BIA indices was evaluated by two obse rvers by comparing a series of 40 measurements of R and Xc and 36 meas urements for weight in four malnourished children on 10 different days during the course of their nutritional recuperation. For weight, the CV (coefficient of variation) was 3.5% for observer A and 3.8% for obs erver B. There were no differences in the means across observers with respect to weight, R, and Xc. The results obtained in this study illus trate that properly trained observers following the same standards can produce equivalent data for BIA and weight measurements even in sick children.