Comparison of body fat percentage assessments by bioelectrical impedance analysis, anthropometrical prediction equations, and dual-energy X-ray absorptiometry in older women

Background: Individuals with high body fat have a higher risk of mortality. Numerous anthropometric-based predictive equations are available for body composition assessments; furthermore, bioelectrical impedance analysis (BIA) estimates are available. However, in older adults, the validity of body fat estimates requires further investigation. Objective: To assess the agreement between percentage body fat (BF%) estimates by BIA and five predictive equations based on anthropometric characteristics using dual X-ray absorptiometry (DXA) as reference method. A secondary objective was to identify whether excluding short-stature women improves the agreement of BF% estimates in a group of community-dwelling, older Mexican women. Methods: -test comparisons and standard error of estimates (SEE) were obtained. The Bland-Altman plot with 95% limits of agreement and the concordance correlation coefficient (CCC) were used to evaluate the BF% prediction equations and BIA estimates. Results: The mean age of the study participants was 73.7 (±5.8) years old. BIA and the anthropometric based equations examined showed mean significant differences when tested in the entire sample. For the taller women (height > 145 cm), no significant difference in the paired comparison was found between DXA and BIA of BF% estimates. The mean BF% was 40.3 (±4.8) and 40.7 (±6.2) for DXA and BIA, respectively. The concordance between methods was good (CCC 0.814), (SEE 2.62). Also, in the taller women subset, the Woolcott equation using waist-to-height ratio presented no significant difference in the paired comparison; however, the error of the estimates was high (SEE 3.37) and the concordance was moderate (CCC 0.693). Conclusion: This study found that BIA yielded good results in the estimation of BF% among women with heights over 145 cm. Also, in this group, the Woolcott predictive equation based on waist circumference and height ratio showed no significant differences compared to DXA in the paired comparison; however, the large error of estimates observed may limit its application. In older women, short stature may impact the validity of the body fat percentage estimates of anthropometric-based predictive equations.