Comparing body composition techniques against an adapted multicompartment model in individuals with excess body...

Accurate body composition assessment is critical for detecting individuals at increased health risk from excess adiposity; however, many measurement techniques lose accuracy in those with a higher body mass index (BMI). This study evaluated the accuracy of body composition techniques against a 4-compartment (4C) model in individuals with overweight or obesity.

N = 75 participants were categorized as having overweight (n = 56, BMI 25-29.9 kg/m²) or obesity (n = 19, BMI ≥30 kg/m²). Body composition was assessed by bioelectrical impedance analysis (BIA), air-displacement plethysmography (ADP), and dual-energy X-ray absorptiometry (DXA). An adapted 4C model used body mass, body volume (via ADP), bone mineral content (via DXA), and total body water (TBW, via BIA). Accuracy was assessed as mean differences (MD) ± standard deviation (comparator - 4C) and 95% limits of agreement (LoA).

ADP demonstrated the smallest overall difference in body fat percentage (BF%; MD = 0.10% ± 1.70%, LoA [-3.23, 3.43], P = 0.620), but its accuracy reduced in individuals with obesity. Both BIA (MD = 1.73% ± 1.72%, LoA [-1.52, 4.98]) and DXA (MD = 1.86% ± 1.79%, LoA [-1.65, 5.37]) overestimated BF% (both P < 0.001).

Overall, ADP demonstrated the best accuracy, while DXA had the greatest differences compared to the 4C model. This may have resulted from TBW being overestimated by BIA, which would inflate 4C-derived fat-free mass. Although all methods demonstrated strong agreement for group-level BF%, a substantial individual-level variability (up to 5% error) highlights the need for caution when interpreting results in clinical or personalized assessment contexts in individuals with excess body weight.