Monash University, Baker IDI Bright Sparks Foundation, Australian Postgraduate Award, VicHealth, National Health and Medical Research Council, Australian Research Council, Victorian Government.

Recent obesity trends in children and adolescents suggest a plateau. However, it is unclear whether such trends have been experienced across socioeconomic groups. We analysed whether recent trends in child and adolescent overweight and obesity differ by socioeconomic position (SEP) across economically advanced countries. Eligible studies reported overweight and obesity prevalence in children and/or adolescents (2-18 years), for at least two time points since 1990, stratified by SEP. Socioeconomic differences in trends in child and adolescent overweight and obesity over time were analysed. Differences in trends between SEP groups were observed across a majority of studies. Over half the studies indicated increasing prevalence among low SEP children and adolescents compared to a third of studies among children and adolescents with a high SEP. Around half the studies indicated widening socioeconomic inequalities in overweight and obesity. Since 2000 a majority of studies demonstrated no change or a decrease in prevalence among both high and low SEP groups. However around 40% of studies indicated widening of socioeconomic inequalities post-2000. While our study provides grounds for optimism, socioeconomic inequalities in overweight and obesity continue to widen. These findings highlight the need for greater consideration of different population groups when implementing obesity interventions.

This study examined the relationship between normal weight, overweight and obesity class I and II+, and the risk of disability, which is defined as impairment in activities of daily living (ADL). Systematic searching of the literature identified eight cross-sectional studies and four longitudinal studies that were comparable for meta-analysis. An additional four cross-sectional studies and one longitudinal study were included for qualitative review. Results from the meta-analysis of cross-sectional studies revealed a graded increase in the risk of ADL limitations from overweight (1.04, 95% confidence interval [CI] 1.00-1.08), class I obesity (1.16, 95% CI 1.11-1.21) and class II+ obesity (1.76, 95% CI 1.28-2.41), relative to normal weight. Meta-analyses of longitudinal studies revealed a similar graded relationship; however, the magnitude of this relationship was slightly greater for all body mass index categories. Qualitative analysis of studies that met the inclusion criteria but were not compatible for meta-analysis supported the pooled results. No studies identified met all of the pre-defined quality criteria, and subgroup analysis was inhibited due to insufficient comparable studies. We conclude that increasing body weight increases the risk of disability in a graded manner, but also emphasize the need for additional studies using contemporary longitudinal cohorts with large numbers of obese class III individuals, a range of ages and with measured height and weight, and incident ADL questions.