Predictive equations for assessing body composition in paediatric population using DXA as the reference method: A systematic review

Suan, W.B. and Jia, L.J. (2021) Predictive equations for assessing body composition in paediatric population using DXA as the reference method: A systematic review. In: Asia Oceania Conference on Obesity & Malaysian Association for the Study of Obesity Scientific Conference 2021, 06-08 Apr 2021, Virtual, Online.

Text FH03-FSK-21-52908.pdf Restricted to Registered users only Download (224kB)

Abstract

Introduction: A simple and practical method such as bioelectrical impedance analysis (BIA) and the anthropometric predictive equation is important to assess the body compositor) of pediatrics for the prevention and management of childhood obesity. Objective: To review the WA- and anthropometric-based predictive equations for assessing body composition in pediatric populations using dual -Energy X-ray absorptiometry IDXA) as the reference method from observational studies Methodology: Comprehensive search of English articles was conducted from electronic databases of PubMed (2005 2020), Scopus (2005-2020). Search terms included predictive equation, bioelectrical impedance analysts (BIA), anthropometric, skinfold, body composition, fat mass, fat-free mass, pediatric populations, children, adolescents, and dual-energy x-ray absorptiometry (DXA). The included studies were those that use WA and/or anthropometry as predictor variables, relative or absolute fat mass (FM) or fat-free mass (FFM) were assessed, and using DXA as the reference method for the healthy pediatric population aged between 2-21 years old. Data extraction was conducted by researcher using predefined data fields, followed by study quality assessment using the Effective Public Health Practice protect (ERHP Ft) Results: A total of 16 studies (25 equations) were included, with the majority of the equations developed from western countries, cross-sectional study design, and with a moderate rating of study quality. Development of predictive equations produces R2 ranges from 0.64 to 0.99, with a low level of RMSE and CV Only 8 out of 16 studies were cross-validated which have high RI ranges from 0.80 to 0.99, with reasonable value for PE, ICC, bias, and limits of agreement. Conclusion: In conclusion, all of the included BIA- and anthropometric predictive equation appears to be acceptable and valid to be used as an alternate method of DXA in the current clinical and held settings. Nevertheless, more validation of predictive equation needs to be done especially in Asian countries.

Actions (login required)

View Item