Benjamin Aretz , University of Rostock, University of Groningen
Gabriele Doblhammer, University of Rostock
Fanny Janssen, Netherlands Interdisciplinary Demographic Institute (NIDI) and University of Groningen
BACKGROUND: Obesity prevalence has globally more than doubled since 1980, whereby 55%-80% of the worldwide increase was contributed by rural regions. However, previous research on the rela-tionship between food infrastructure and obesity did not regard effect differences between urban and rural areas. Our study investigated this relationship using spatial analysis stratified by area of residence (urban vs rural). DATA AND METHODS: We used data on obesity prevalence among Dutch aged 19+ and on distance to food infra-structure by neighbourhoods in 2016. We distinguished two food infrastructure domains i) cafeterias and fast food stores and ii) grocery and fresh food stores. We analysed 2,717 (1,093 urban,1,624 rural) neighbourhoods. First, we mapped the geographical difference in obesity. Second, multivariable spatial error models stratified for urban and non-urban regions were estimated. Third, geographically weighted regressions (GWR) were applied to explore spatial effect variation. RESULTS: Average obesity prevalence was a bit higher in urban (14.61%) than in non-urban neighbour-hoods (14.14%). Higher distance to both food infrastructure domains were positively associ-ated with obesity in urban and negatively in rural neighbourhoods. In our GWRs, we replicat-ed the results from the spatial lag models. That is, the urban/rural pattern became much clearer when controlling for population composition and lifestyle confounders. CONCLUSION: Our findings confirm previous findings from the US that there are different spatial patterns for the relationship between food infrastructure and obesity in urban and rural spaces, which for the Netherlands seem to be explained by causal effects in rural and selection effects in urban regions.
Presented in Session P2. Poster Session Ageing, Health and Mortality