Jeroen de Bont1, Sandra Márquez2, Sílvia Fernández-Barrés2, Charline Warembourg2, Sarah Koch2, Cecilia Persavento2, Silvia Fochs2, Núria Pey2, Montserrat de Castro2, Serena Fossati2, Mark Nieuwenhuijsen2, Xavier Basagaña2, Maribel Casas2, Talita Duarte-Salles3, Martine Vrijheid4. 1. ISGlobal, Barcelona, Spain; Spanish Consortium for Research on Epidemiology and Public Health (CIBERESP), Spain; Universitat Pompeu Fabra, Barcelona, Spain; Fundació Institut Universitari per a la recerca a l'Atenció Primària de Salut Jordi Gol i Gurina (IDIAPJGol), Barcelona, Spain; Universitat Autònoma de Barcelona, Bellaterra (Cerdanyola del Vallès), Spain. 2. ISGlobal, Barcelona, Spain; Spanish Consortium for Research on Epidemiology and Public Health (CIBERESP), Spain; Universitat Pompeu Fabra, Barcelona, Spain. 3. Fundació Institut Universitari per a la recerca a l'Atenció Primària de Salut Jordi Gol i Gurina (IDIAPJGol), Barcelona, Spain. 4. ISGlobal, Barcelona, Spain; Spanish Consortium for Research on Epidemiology and Public Health (CIBERESP), Spain; Universitat Pompeu Fabra, Barcelona, Spain. Electronic address: martine.vrijheid@isglobal.org.
Abstract
BACKGROUND: Urban environments are characterised by many factors that may influence children's lifestyle and increase the risk of childhood obesity, but multiple urban exposures have scarcely been studied. OBJECTIVE: We evaluated the association between multiple urban exposures and childhood obesity outcomes and weight-related behaviours. METHODS: We conducted a cross-sectional study including 2213 children aged 9-12 years in Sabadell, Spain. We estimated ambient air pollution, green spaces, built and food environment, road traffic and road traffic noise at residential addresses through a total of 28 exposure variables in various buffers. Childhood obesity outcomes included body mass index (BMI), waist circumference and body fat. Weight-related behaviours included diet (fast food and sugar-sweetened beverage consumption), physical activity, sedentary behaviour, sleep duration and well-being. Associations between exposures (urban environment) and outcomes (obesity and behaviours) were estimated in single and multiple-exposure regression models and in a hierarchical clustering on principal components (HCPC) analysis. RESULTS: Forty percent of children were overweight or obese. In single exposure models, very few associations were observed between the urban exposures and obesity outcomes or weight-related behaviours after correction for multiple testing. In multiple exposure models, PMcoarse, denser unhealthy food environment and land use mix were statistically significant associated with childhood obesity outcomes (e.g 17.7 facilities/km2 increase of unhealthy food environment (OR overweight/obesity status) = 1.20 [95% CI: 1.01; 1.44]). Cluster analysis identified 5 clusters of urban exposures. Compared to the most neutral cluster, the cluster with high air pollution, road traffic, and road noise levels was associated with a higher BMI and higher odds of overweight and obesity (β (zBMI) = 0.17, [95% CI: 0.01, 0.17]; OR (overweight/obesity) = 1.36, [95% CI: 0.99, 1.85]); the clusters were not associated with the weight-related behaviours. CONCLUSIONS: This systematic study of many exposures in the urban environment suggests that an exposure pattern characterised by higher levels of ambient air pollution, road traffic and road traffic noise is associated with increased childhood obesity risk and that PMcoarse, land use mix and food environment are separately associated with obesity risk. These findings require follow-up in longitudinal studies and different settings.
BACKGROUND: Urban environments are characterised by many factors that may influence children's lifestyle and increase the risk of childhood obesity, but multiple urban exposures have scarcely been studied. OBJECTIVE: We evaluated the association between multiple urban exposures and childhood obesity outcomes and weight-related behaviours. METHODS: We conducted a cross-sectional study including 2213 children aged 9-12 years in Sabadell, Spain. We estimated ambient air pollution, green spaces, built and food environment, road traffic and road traffic noise at residential addresses through a total of 28 exposure variables in various buffers. Childhood obesity outcomes included body mass index (BMI), waist circumference and body fat. Weight-related behaviours included diet (fast food and sugar-sweetened beverage consumption), physical activity, sedentary behaviour, sleep duration and well-being. Associations between exposures (urban environment) and outcomes (obesity and behaviours) were estimated in single and multiple-exposure regression models and in a hierarchical clustering on principal components (HCPC) analysis. RESULTS: Forty percent of children were overweight or obese. In single exposure models, very few associations were observed between the urban exposures and obesity outcomes or weight-related behaviours after correction for multiple testing. In multiple exposure models, PMcoarse, denser unhealthy food environment and land use mix were statistically significant associated with childhood obesity outcomes (e.g 17.7 facilities/km2 increase of unhealthy food environment (OR overweight/obesity status) = 1.20 [95% CI: 1.01; 1.44]). Cluster analysis identified 5 clusters of urban exposures. Compared to the most neutral cluster, the cluster with high air pollution, road traffic, and road noise levels was associated with a higher BMI and higher odds of overweight and obesity (β (zBMI) = 0.17, [95% CI: 0.01, 0.17]; OR (overweight/obesity) = 1.36, [95% CI: 0.99, 1.85]); the clusters were not associated with the weight-related behaviours. CONCLUSIONS: This systematic study of many exposures in the urban environment suggests that an exposure pattern characterised by higher levels of ambient air pollution, road traffic and road traffic noise is associated with increased childhood obesity risk and that PMcoarse, land use mix and food environment are separately associated with obesity risk. These findings require follow-up in longitudinal studies and different settings.