Associations between children's diets and features of their residential and school neighbourhood food environments.

OBJECTIVES: Among studies of the built environment, few examine neighbourhood food environments in relation to children's diets. We examined the associations of residential and school neighbourhood access to different types of food establishments with children's diets.

METHODS: Data from QUALITY (Quebec Adipose and Lifestyle Investigation in Youth), an ongoing study on the natural history of obesity in 630 Quebec youth aged 8-10 years with a parental history of obesity, were analyzed (n=512). Three 24-hour diet recalls were used to assess dietary intake of vegetables and fruit, and sugar-sweetened beverages. Questionnaires were used to determine the frequency of eating/snacking out and consumption of delivered/take-out foods. We characterized residential and school neighbourhood food environments by means of a Geographic Information System. Variables included distance to the nearest supermarket, fast-food restaurant and convenience store, and densities of each food establishment type computed for 1 km network buffers around each child's residence and school. Retail Food Environment indices were also computed. Multivariable logistic regressions (residential access) and generalized estimating equations (school access) were used for analysis.

RESULTS: Residential and school neighbourhood access to supermarkets was not associated with children's diets. Residing in neighbourhoods with lower access to fast-food restaurants and convenience stores was associated with a lower likelihood of eating and snacking out. Children attending schools in neighbourhoods with a higher number of unhealthful relative to healthful food establishments scored most poorly on dietary outcomes.

CONCLUSIONS: Further investigations are needed to inform policies aimed at shaping neighbourhood-level food purchasing opportunities, particularly for access to fast-food restaurants and convenience stores.