Abstract
The Body Adiposity Index (BAI) is a practical anthropometric method used to measure body fat (BF) percentage (BF%). Recently developed, the validity and precision of BAI has been studied with adult samples of men and women, populations from different countries and ethnicities, varying amounts of BF, and sensitivity to detecting change over time. However, it is still necessary to determine its potential use in clinical practice and epidemiologic studies. Thus, our objective was to verify, through a systematic review, the validity of the BAI in predicting BF% in adults. Two independent researchers performed a search using PubMed, Web of Science, Science Direct, and Scopus databases. In order to be included, the studies had to use dual-energy X-ray absorptiometry (DXA) as a reference method. We excluded studies with samples from individuals with diseases or syndromes that alter the regional distribution of BF%. We included 19 studies with samples on individuals from different continents, varied ethnicities, both sexes, and a wide age range (18-83 y). The concordance of the BAI with DXA assessed by Lin's concordance correlation coefficient showed results classified as poor (pc < 0.90). Bland-Altman plots showed that the BAI produced large individual errors when predicting BF% in all studies using this analysis. The studies were consistent in affirming that the BAI showed limited capacity to estimate BF% in adults. The BAI shows wide individual errors, in agreement with the reference method, and a lack of sensitivity in detecting change in BF% over time. The method presents a systematic error of BF% overestimation in individuals with ≤20% of BF, and underestimation in individuals with >30% of BF, regardless of sex, age, and ethnicity. The results of this systematic review show enough evidence that the BAI does not present satisfying results, and its use is not recommended for BF% determination in adults.