Body composition models based on anthropometry: systematic literature review

The body composition determination is carried out through measurement methods that require the use of specialized equipment that is difficult to acquire and manipulate. Therefore, different authors have developed mathematical models for its calculation. The target of this review was to analyze the work carried out on mathematical models of body composition variables from different anthropometric measurements, answering the following questions: what is the corporal variable that the model predicts?, what are the input variables for model development?, how is the patients typification in each model?, what data analysis method has been used?, and how has been the model evaluated? The search was limited to journals found in repositories in the areas of Medicine, Nursing, Biochemistry, Biology, Health, Pharmacology, Immunology, Engineering, and Mathematics. Four hundred and twenty-four articles were found, which were reduced to 30 by applying the systematic literature review process. The analyzed studies are oriented to the prediction of variables related to body fat mass. The evaluation results found for fat-free mass, fat mass and metabolic rate differ according to the comparison technique and the body segments analyzed. The evaluation is mainly based on the intraclass correlation, the Pearson correlation and the coefficient of determination (r(2)), and they present a good correlation for the population under study.