Related Experiment Video
Updated: Jun 24, 2026
Assessment of Child Anthropometry in a Large Epidemiologic Study
Published on: February 2, 2017
Body fat measured by bioelectrical impedance in Hong Kong Chinese children
Rita Y T Sung1, H K So, K C Choi
1Department of Paediatrics, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong. yntzsung@suhk.edu.hk
Objective:
To establish reference standards for percentage body fat measured in Hong Kong Chinese children, by methods involving bioelectrical impedance analysis.
Design:
Cross-sectional study.
Setting:
Thirty-six randomly selected primary and secondary schools and a teaching hospital in Hong Kong.
Participants:
A total of 14 842 students randomly selected from the schools and two additional small convenience samples of subjects.
Main Outcome Measures:
Percentage body fat was measured with the Tanita Body Composition Analyzer (Model no. BF-522) and percentile curves were constructed using the LMS method. In one separate small sample of children, repeatability of the percentage body fat measurement was assessed at different times of the day by BF-522 bioelectrical impedance analysis. In another sample, assessment was by the BF-522 and two other models (BC-418 and BF-401) consecutively to test the agreement of percentage body fat values obtained by the three different models. RESULTS. The percentage body fat values and percentile curves are presented. From the age of 6 to 18 years, the percentage body fat remained fairly stable in boys, but increased steadily in girls. The mean difference in percentage body fat measured with BF-522 at different times of the day was around 1% (95% limits of agreement: -4% to +8%). The mean differences in readings obtained from the BC-418 and BF-522 devices were -3.5% and 1% in boys and girls, respectively. The 95% limits of agreement were particularly wide in boys (-15% to 8%).
Conclusions:
Reference values for percentage body fat of Chinese children and adolescents are provided. Caution needs to be exercised however, given that readings obtained at different times of the day vary and data obtained by different makes and models of bioelectrical impedance analysis machines may not be interchangeable.
