Seasonal variation in arm girth to height ratio as an alternative to body mass index of Ellisras rural children aged 5–15 Years: Ellisras Longitudinal Study

Abstract

Obesity has been a growing concern globally being the second greatest cause of mortality and morbidity. This study aimed to investigate the seasonal variation of arm girth to height as an alternative to body mass index (BMI) in Ellisras rural children aged 5–15 years. Anthropometric measurements were taken in 1986 (1034 boys and 952 girls) Ellisras children aged 5-15 years. Median values and frequency analysis determined the prevalence of obesity using BMI and arm girth to height ratio (AHtR). The receiver operating characteristic (ROC) curve analysis was used to determine the best predictor of obesity between BMI and AHtR. Girls aged 5–10 years had a median arm girth of 16.7 cm in spring compared to 16.5 cm in autumn. However, boys of the same age had no median seasonal variation in arm girth, height and weight. Girls aged 11–15 years had a median weight of 32.5 kg in spring and 28 kg in autumn. In addition, there was a significant median difference (p<0.05) in weight among girls and boys of the same age. The prevalence of obesity by arm girth in girls aged 5–10 was higher in autumn (5.0%) compared to spring (3.7%), and boys had a higher prevalence in spring (2.4%) than in autumn (2.2%). The overall prevalence of obesity was 5.1% when it was assessed by both BMI and AHtR for the cohort. Obesity assessed by BMI had 100% sensitivity and specificity in boys and girls, whereas when evaluated by AHtR it had a 90.2% sensitivity and 81.2% specificity in boys; 80.3% sensitivity and 83.1% specificity in girls (p < 0.05). There was seasonal variation in BMI and AHtR among the Ellisras rural children. This study showed BMI to be less affected by gender, but a better predictor of obesity than AHtR. However, AHtR could be used as an alternative tool to BMI in evaluating obesity in the children.