Impact of ethnicity upon body composition assessment in Sri Lankan Australian children

• Published in: Journal of paediatrics and child health Vol. 41; no. 3; pp. 101 - 106
• Main Authors: Wickramasinghe, VP, Cleghorn, GJ, Edmiston, KA, Davies, PSW
• Format: Journal Article
• Language: English
• Published: PO Box 378 , Carlton South Victoria 3053 , Australia Blackwell Science Pty 01.03.2005; Blackwell Publishing Ltd
• Subjects: Adolescent; Age Distribution; Anthropometry; Australia; Australia - epidemiology; bioelectrical impedance; Body Composition; Body fat; Body Mass Index; Child; Children; Electric Impedance; Female; Humans; Male; Obesity; Obesity - ethnology; Sex Distribution; Sri Lanka - ethnology; Sri Lankan Australian children; Sri Lankan people; Australia; Sri Lanka
• ISSN: 1034-4810; 1440-1754
• DOI: 10.1111/j.1440-1754.2005.00558.x

Objectives:  Obesity is a disease with excess body fat where health is adversely affected. Therefore it is prudent to make the diagnosis of obesity based on the measure of percentage body fat. Body composition of a group of Australian children of Sri Lankan origin were studied to evaluate the applicability of some bedside techniques in the measurement of percentage body fat. Methods:  Height (H) and weight (W) was measured and BMI (W/H2) calculated. Bioelectrical impedance analysis (BIA) was measured using tetra polar technique with an 800 μA current of 50 Hz frequency. Total body water was used as a reference method and was determined by deuterium dilution and fat free mass and hence fat mass (FM) derived using age and gender specific constants. Percentage FM was estimated using four predictive equations, which used BIA and anthropometric measurements. Results:  Twenty‐seven boys and 15 girls were studied with mean ages being 9.1 years and 9.6 years, respectively. Girls had a significantly higher FM compared to boys. The mean percentage FM of boys (22.9 ± 8.7%) was higher than the limit for obesity and for girls (29.0 ± 6.0%) it was just below the cut‐off. BMI was comparatively low. All but BIA equation in boys under estimated the percentage FM. The impedance index and weight showed a strong association with total body water (r2= 0.96, P < 0.001). Except for BIA in boys all other techniques under diagnosed obesity. Conclusions:  Sri Lankan Australian children appear to have a high percentage of fat with a low BMI and some of the available indirect techniques are not helpful in the assessment of body composition. Therefore ethnic and/or population specific predictive equations have to be developed for the assessment of body composition, especially in a multicultural society using indirect methods such as BIA or anthropometry.