Modified laboratory- and field-based multicompartment models have been found valid for estimating body composition. However, the comparability between these models is unknown.
Purpose: This study determined the validity of field and laboratory three-compartment (3C) models in healthy adults.
Methods: One hundred twenty participants (63 men and 57 women; age, 22 ± 5 yr; BMI, 24.9 ± 3.9 kg·m) participated in this study. A criterion four-compartment model (4C-Criterion) was determined with underwater weighing for body volume (BV), bioimpedance spectroscopy for total body water (TBW), and dual-energy X-ray absorptiometry (DXA) for bone mineral content. Modified laboratory-based 3C models were determined using bioimpedance spectroscopy for TBW and two separate DXA BV equations (3C-DXASR and 3C-DXAW) whereas a field-based 3C model (3C-Field) was obtained using single-frequency bioimpedance analysis for TBW and skinfold-derived BV. In addition, a stand-alone DXA assessment was evaluated.
Results: The effect size of the mean differences when compared to the 4C-Criterion were trivial to small for all modified 3C models and DXA when estimating fat mass, fat-free mass, and body fat percentage. The standard error of estimate and 95% limits of agreement for all modified 3C models and DXA were similar and considered acceptable. However, 3C-Field produced the lowest total error values and 3C-DXASR produced slightly lower total error values than 3C-DXAW and DXA.
Conclusions: The present study found that all modified 3C models and DXA exhibited acceptable errors. When performed by expert personnel, a field-based 3C model appears to be a viable alternative to laboratory-derived models in healthy adults.