Factors related to resting energy expenditure and physical activity of 6–9-year old children in two primary schools in the City of Tshwane metropolitan area

Abstract

A lower resting energy expenditure (REE) has been suggested to partially explain the disproportionate prevalence of overweight/obesity among black African women, yet no studies have investigated the REE of South African (SA) children. Similarly, physical activity (PA) as a modifiable factor related to childhood energy expenditure is under-researched in the local context. The study determined the relationship between sex and population group (determinant factors), on the one hand, and REE and PA (outcomes) of 6–9-year-old SA children attending two primary schools in the City of Tshwane metropolitan area, on the other, taking phenotypic characteristics as confounders (mediating factors) and the study context into account. In a cross-sectional study with quota sampling, the REE of 6–9-year-old children attending two urban schools in SA was measured with indirect calorimetry (IC), and PA with a pedometer. Multifrequency bioelectrical impedance analysis was used to assess body composition (BC) (fat-free mass [FFM], FFM index, fat mass [FM] and FM index). Multivariate regression was used to calculate REE and PA adjusted for phenotypic confounders (z-scores of weight-for-age, height-for-age and body mass index-for-age, and BC). Sex and population differences in REE and PA (measured and adjusted) were determined with two-way ANOVA. Ninety-four healthy children (59.6% girls; 52.1% black African) with similar socio-economic status and access to PA participated in the study. Despite variations in BC, sex differences in REE were not significant (41 kcal/day ≈ 172 kJ/day; P = 0.375). The REE in black African participants was significantly lower than in their white counterparts (146 kcal/day ≈ 613 kJ/day; P = 0.002). When adjusting for BC, population differences in REE declined, especially after adjustment for FFM (91 kcal/day ≈ 382 kJ/day; P = 0.039), but remained clinically significant. Average steps/day in girls (10212 [9519;10906]) was lower than in boys (11433 [10588;12277]) (P = 0.029), and lower in black African (9280 [8538;10022]) than in white (12258 [11483;13033]) (P < 0.001) participants. No significant relationship (r = 0.05; P = 0.651) was observed between REE and PA. Within the context of a similar SES and PA environment, the REE and PA of black African children was lower than white. Differences in REE between sexes were not significant, but girls had a lower PA than boys.