P64 Exploring weight status in Australia and the US: a cross-sectional analysis using a commercial geodemographic classification

Abstract

Background: Obesity prevalence of epidemic proportions continues to be a major public health problem globally. Better understanding of the spatial and social variation in obesity is essential in order to support changes to policy or our environment to reduce obesity prevalence. This paper uses a geodemographic classification – which combines demographic characteristics with a small area geographic unit – to profile weight status and estimate small-area obesity prevalence in Australia and the US. Methods: This study is a cross sectional analysis of two large studies; the Australian Longitudinal Study on Women’s Health (ALSWH) and the Seattle Obesity Study (SOS1). Descriptive statistics, chi2 and Kruskal Wallis test for difference, linear and multinomial logistic regression were carried out using Stata 12 statistical software. ArcMap10 was used to: (1) match the study participants to a CAMEO geodemographic identifier, using the longitude and latitude of their home address and (2) to visualise the obesity estimates for Newcastle (Australia) and Seattle (US). CAMEO is a commercially available geodemographic classification. Results Both ALSWH and SOS1 had under and over-representation in certain CAMEO groups compared to national representation, but each group contains high numbers of individuals suggesting results are robust. Demographic characteristics of the study participants are in line with those expected in the corresponding CAMEO groups. In both studies significant differences in body mass index across CAMEO groups exists (p < 0.001). In Australia, the Diverse Low Income Urban Communities had twice the odds of being obese than the Affluent Urban Professionals (OR = 2.24 (95% CI 1.55 to 3.23)). In the US, compared to the American Aristocracy, the Enterprising Households (OR 1.97 (95% CI 1.25 to 3.09)), Comfortable Communities (OR 2.01 (95% CI 1.25 to 3.22)) and Dynamic Neighbourhoods (2.09 (1.30 to 3.36)) also had twice the odds of being obese. Comprehensive obesity maps for Newcastle and Seattle at a small area geographical resolution were produced, identifying neighbourhoods with likely high prevalence of obesity. Conclusion: Geodemographic classifications, such as CAMEO, combined with survey data offer promising solutions for profiling obesity outcomes worldwide which could facilitate effective targeting of potential public health interventions at a neighbourhood geography scale.