ABSTRACT
OBJECTIVES: The prevalence of cardiometabolic disease has risen in Africa and parallels the obesity epidemic. To assess cardiometabolic risk, body composition measurements by dual-energy X-ray absorptiometry (DXA) are ideal. In communities with limited resources, alternative measures may be useful but have not been compared extensively in black Africans. Therefore, the aim of this study was to identify alternative methods of body composition assessment, such as body adiposity index (BAI) and bioelectrical impedance analysis (BIA), for use in African-born blacks. METHODS: This was a cross-sectional study with African-born blacks. BAI and five BIA predictive equations (using variations of height, weight, age, sex, and impedance) were compared with DXA to estimate percent fat. Participants were 266 African-born blacks (39 ± 10 y, body mass index 28 ± 4 kg/m2, and 68% men) living in metropolitan Washington DC. Equivalence (90% confidence interval, -3 to 3), concordance, and Bland-Altman analyses (bias <2%, R2 closest to zero) compared BAI or BIA predictive equations to DXA as the criterion method. RESULTS: DXA percent fat was 27.2% ± 5.5% and 40.3% ± 6.9% in men and women, respectively. BAI underestimated percent fat in men (bias: 1.88 ± 4.71, R2 = 0.25, P < 0.001) and women (bias: 6.47 ± 4.94, R2 = 0.08, P = 0.01). Of the five BIA predictive equations, the equation reported by Sun et al. had the best agreement with DXA percent fat for men (bias: -0.91 ± 3.67, R2 = 0.02, P = 0.05) and women (bias: -0.92 ± 4.02, R2 = 0.003, P = 0.58). Percent fat from the Sun et al. equation best agreed with DXA percent fat. CONCLUSION: BIA with the Sun et al. predictive equation was the best alternative to DXA for body fat assessment in African-born blacks.
