Agreement Between Standing Eight-Point Multifrequency Bioelectrical Impedance Analysis and Dual-Energy X-Ray...

Background/Objectives: Multifrequency bioelectrical impedance analysis (MF-BIA) is increasingly used for practical body composition assessment when dual-energy X-ray absorptiometry (DXA) is unavailable or impractical. However, MF-BIA estimates are device-, population-, and outcome-specific, and therefore require validation against reference methods under standardized conditions. This study examined the agreement, concordance, and systematic bias between a standing 8-point MF-BIA device and DXA-derived body composition estimates in apparently healthy Greek adults. Methods: A total of 1250 adults aged 18 to 80 years completed same-day DXA and MF-BIA (Charder MA801) assessments. Fat mass (FM), fat-free mass (FFM), body fat percentage (BF%), and appendicular skeletal muscle mass estimate (ASM) were compared between methods. Analyses were performed by sex and BMI category. Pearson correlations described association, whereas Bland-Altman analysis, Lin's concordance correlation coefficient (CCC), mean absolute error (MAE), root mean square error (RMSE), and proportional bias testing evaluated agreement and error magnitude. Results: MF-BIA showed strong associations with DXA-derived outcomes, but systematic bias was observed. When BMI categories were considered collectively, MF-BIA underestimated BF% by 3.59 percentage points in men and 4.25 percentage points in women, underestimated FM by 2.89 kg and 2.58 kg, and overestimated FFM by 3.09 kg and 3.29 kg, respectively. CCC was highest for FM (men: 0.913; women: 0.949) and lower for FFM and ASM in women (0.642 and 0.714, respectively). Proportional bias was observed for BF%, FM, and ASM in both sexes, and for FFM in women. Conclusions: The MA801 showed strong associations and outcome-specific concordance with DXA, but systematic bias and individual-level error limit interchangeability. Under standardized conditions, MF-BIA may support group-level or repeated same-device assessments but not precise individual-level assessment, clinical classification, or monitoring of small longitudinal changes.