• Mariana De Santis Filgueiras, Roberta Stofeles Cecon, Eliane Rodrigues de Faria, Franciane Rocha de Faria, Patrícia Feliciano Pereira, Andréia Queiroz Ribeiro, Silvia Eloiza Priore, and Juliana Farias de Novaes.
• 1Department of Nutrition and Health,Biological Sciences Center,Federal University of Viçosa,CEP 36570-900, Viçosa, MG,Brazil.
• Public Health Nutr. 2019 Jan 1; 22 (1): 132-139.

ObjectiveTo evaluate agreement of the body adiposity index (BAI) and paediatric body adiposity index (BAIp) in estimating body fat compared with dual-energy X-ray absorptiometry (DXA) and to propose cut-off points for these indices to classify excess adiposity in Brazilian children and adolescents.DesignCross-sectional study. Measures of weight, height, hip circumference, BMI and body fat percentage (%BF) assessed by DXA were taken, and BAI and BAIp were calculated. The Bland-Altman plot was used to estimate agreement between the methods, and the receiver-operating characteristic curve to determine the cut-off points for BAI and BAIp per age and sex in comparison with DXA.SettingViçosa, Minas Gerais, Brazil.SubjectsChildren and adolescents aged 8-19 years (n 1049).ResultsOf the children and adolescents, 52·4 % were girls. BAI and BAIp had satisfactory performance by the receiver-operating characteristic curve, except for the 18-19 years age group, whose BAIp had better predictive capacity than BAI. The agreement analysis showed that BAI overestimated %BF by 2·64 %, on average, using DXA; while BAIp underestimated %BF by 3·37 %.ConclusionsBAI and BAIp showed low agreement with the body fat obtained by DXA, requiring caution when interpreting body composition data in children and adolescents.

28 keywords...

hide…