Relative accuracy of anthropometric-based body fat equations in males and females with varying BMI classifications

Background: BMI based body fat equations developed from Womersley and Durnin (BMIWO), Jackson et al. (BMIJA), Deurenberg et al. (BMIDE), and Gallagher et al. (BMIGA) are commonly used to quantify body fat percentage (BF%). However, relative fat mass (RFM) is a new anthropometric-based method that has been proposed as an alternative. Aims: The purpose of this study was to examine the independent and interactive effects of sex and BMI classification on the relative accuracy of BMI-based body fat equations and RFM. Methods: Males (n = 75) and females (n = 75) were stratified and classified into three different groups; 1) normal weight (n = 50 [NW: 50% males]; BMI<25.0 kg/m(2)); 2) overweight (n = 50 [OW: 50% males]; BMI >= 25.0-29.9 kg/m(2)); 3) obese (n = 50 [OB: 50% males]; BMI >= 30.0 kg/m(2)). A criterion three-compartment model (3C model) was determined with air displacement plethysmography for body volume and multi-frequency bioimpedance analysis for total body water. Data were stratified by sex and BMI classification. Difference scores were created by subtracting estimated BF% from 3C model BF%. Results: A significant SEX x BMI interaction was detected for all comparisons (all p < 0.05). Post hoc analysis indicated the differences in BF% were statistically significant between OW females and males for all equations (BMIWO:-2.99 +/- 4.79% vs. 4.71 +/- 5.86%, p = 0.003; BMIJA:-1.77 +/- 4.83% vs. 5.77 +/- 5.85%, p < 0.001; BMIDE:-3.09 +/- 4.80% vs. 4.97 +/- 5.98%, p < 0.001; BMIGA:0.36 +/- 4.51% vs. 4.56 +/- 5.55%, p = 0.018; RFM:-2.17 +/- 4.84% vs. 3.01 +/- 5.34%, p = 0.004, respectively). In addition, there were significant differences between females and males classified as NW(BMIJA:-2.11 +/- 4.15% vs. 2.61 +/- 5.98%, p = 0.008) and OB (BMIGA:2.40 +/- 3.36% vs. -1.09 +/- 6.40%, p = 0.006). Conclusions: The current findings highlight that RFM does not appear to overcome error commonly associated with BMI-based body fat equations when stratifying by sex and BMI classification. Nonetheless, practitioners can use BMIWO, BMIDE, and RFM in males and females classified as NW or OB, but should employ caution prior to use in OW persons. (c) 2019 European Society for Clinical Nutrition and Metabolism. Published by Elsevier Ltd. All rights reserved.