Obesity prevalence in a sample of low socioeconomic children: Cross sectional associations with physical activity, sedentary and eating behaviour

Abstract

Background: Childhood obesity prevalence, previously at a disturbing upward trend, appears to be holding at a steady rate. The purpose of this study was to investigate current levels of obesity prevalence in a sample of low socioeconomic preadolescent children and examine associations with physical activity, sedentary and eating behaviour. Methods: Nine primary schools in Victoria, Australia were randomly selected from the bottom two strata of a five level socio-economic scale. Self-report physical activity, screen-time and eating behavioural data were collected from students (age 10—13) via questionnaire (N = 310). Physical activity intensity and duration data were collected via Actigraph GT1M accelerometer. Anthropometric (height, weight and waist) measurements were used to calculate BMI-z. Descriptive analyses by gender were assessed using chi square and t-tests as appropriate. Linear regression tested for strength of association between BMI-z and behaviours. Results: Overweight prevalence 37.7% (43% male; 34% female), mean age 12.1 (58% female). Males were more active (M: 2.5 h, F: 1.8 h; Z = 2.92; P < 0.01) and engaged in more screen time (M: 4 h, F: 3 h; Z = 2.36; P < 0.05) on weekends than females. Type of screen behaviour differed by gender, more females used computers for leisure (F: 57%; M: 44%; X2 = 5.32; P < 0.05) andmoremales played nonactive computer games (M: 53%; F: 33%; X2 = 12.55; P < 0.01). Non-core food and sweetened beverage intake target scores were not met by 93% and 58% students respectively. No statistically significant associations were found between each behaviour and BMI-z. Conclusion: The patterns of obesogenic behaviour require different obesity prevention strategies per gender when targeting low socioeconomic pre-adolescent children, as obesity prevalence remains at a high level in this sample.