Graybeal, AJ, Moore, ML, Cruz, MR, and Tinsley, GM. Body composition assessment in male and female bodybuilders: a 4-compartment model comparison of dual-energy x-ray absorptiometry and impedance-based devices. J Strength Cond Res 34(6): 1676-1689, 2020-The purpose of this study was to examine the group and individual accuracy of body composition estimates obtained from multicompartment models, dual-energy x-ray absorptiometry (DXA), and several impedance-based devices in male and female bodybuilders. Twenty-seven male (n = 17; 4-compartment [4C] model fat-free mass index [FFMI]: 25.1 ± 1.8 kg·m; 4C body fat: 11.8 ± 4.4%) and female (n = 10; 4C FFMI: 18.3 ± 1.4 kg·m; 4C body fat: 19.7 ± 4.9%) bodybuilders underwent duplicate assessments using DXA, bioimpedance spectroscopy (BIS), electrical impedance myography (EIM), and 3 bioelectrical impedance analysis (BIA) devices. In addition to utilizing standard output, multicompartment models were generated. For each method, body fat %, FFM, and fat mass were compared with the reference 4C model for the evaluation of group and individual errors. The 3-compartment model with a BIS body water estimate produced the lowest standard error of the estimate, total error (TE), and 95% limits of agreement (LOA) for all variables, although some alternative methods had lower constant error (CE). In general, multicompartment models with BIS or multifrequency BIA body water estimates produced more accurate body composition estimates than single assessment techniques (i.e., DXA, BIS, EIM, and BIA). Single assessment techniques produced low CE and TE for some body composition variables. However, proportional bias was observed for DXA and BIS. All single assessment techniques produced LOA large enough to make the utility of these methods questionable in individual athletes. Appropriate caution should be used when interpreting and utilizing body composition estimates in muscular physique athletes, particularly at the individual level.