Prediction of fat-free mass by bioimpedance analysis in migrant Asian Indian men and women: a cross validation study

Background: Bioimpedance analysis (BIA) is a fast and convenient field technique for estimation of total body fat-free mass (FFM). However, bioimpedance-based prediction equations have been developed in predominantly white populations and little information is available on their usefulness in Asian Indian populations. Objective: To develop a prediction equation for FFM based on BIA measurements applicable to a migrant population of Asian Indians and to investigate the predictive accuracy of published BIA-based equations in this ethnic group. Design: FFM was measured by dual-energy X-ray absorptiometry in 211 healthy adults (110 men, 101 women; age ranges 19-74 year) of Asian Indian ethnicity and used as the reference measure to develop prediction equations based on single-frequency BIA measurements of resistance and reactance. A cross-validation technique was applied. Predictive accuracy of published BIA-based equations was assessed in this sample. Results: Sex-specific equations developed in the entire group included height(2)/resistance and body weight as predictors (R-2=0.84 and 0.70 and standard errors of estimate of 2.8 and 2.0 kg for males and females, respectively; CV: 6%). Of published equations examined, one predicted FFM satisfactorily in men with nonsignificant bias and may be applicable to Asian Indian populations. None of the published equations tested performed satisfactorily in women. Conclusions: Bioimpedance-based equations for predicting FFM developed specifically in Asian Indians are recommended for field studies designed to measure body composition of this ethnic group.