Cross-validation of prediction equations for estimating the body fat percentage in adults with obesity.

Abstract

An accurate estimation of the body fat percentage (BF%) in patients with obesity is of clinical importance. Therefore, we aim to assess the validity of anthropometric-based BF predictive equations in treatment-seeking patients in an outpatient setting. BF% was assessed by Tanita MC-780MA bioimpedance (BIA) and considered as a reference method, and anthropometric-based predictive equations were used in BF% estimations among 275 adults of both genders, in the outpatient clinic of the Department of Nutrition and Dietetics at Beirut Arab University (Lebanon). The mean differences between the measured and estimated BF% values were calculated to assess the accuracy of five equations, and the Bland-Altman method was used to assess the level of agreement. In both males and females, all predictive equations gave significantly different estimates of BF% when compared to those measured by BIA. On the other hand, in both genders, the mean difference between the BF% value estimated by the Jackson equation and that measured using BIA, was not significant, and agreement was confirmed using Bland-Altman plots. We suggest the Jackson equation for accurate BF% estimation in both genders of patients with obesity in the Lebanese population. However, future studies are still urgently needed to develop and validate new predictive equations suitable for BF% estimations, taking into account ethnicity (i.e., the Arab population).