Associations between residential food environment and dietary patterns in urban-dwelling older adults: results from the VoisiNuAge study

• Published in: Public health nutrition 2012-11, Vol.15 (11), p.2026-2039
• Main Authors: Mercille, Geneviève, Richard, Lucie, Gauvin, Lise, Kestens, Yan, Shatenstein, Bryna, Daniel, Mark, Payette, Hélène
• Format: Article
• Language: English
• Subjects: Adults; Aged; Aged, 80 and over; Aging; Canada; Chronic illnesses; Commerce; Cross-Sectional Studies; Diet; Diet - standards; Environment; Fast food; Fast Foods; Female; Food; Food Supply - economics; Geographic information systems; Geriatrics; Humans; Investigations; Linear Models; Male; Metropolitan areas; Monitoring and Surveillance; Neighborhoods; Nutrition research; Older people; Quality of life; Remote sensing; Restaurants; Shopping; Supermarkets; Surveys and Questionnaires; Urban Population
• ISSN: 1368-9800; 1475-2727
• DOI: 10.1017/S136898001200273X

To examine associations between the availability of residential-area food sources and dietary patterns among seniors. Cross-sectional analyses. Individual-level data from the NuAge study on nutrition and healthy ageing were merged with geographic information system data on food store availability and area-level social composition. Two dietary patterns reflecting lower- and higher-quality diets (respectively designated 'western' and 'prudent') were identified from FFQ data. Two food source relative availability measures were calculated for a 500 m road-network buffer around participants' homes: (i) proportion of fast-food outlets (%FFO) relative to all restaurants and (ii) proportion of stores potentially selling healthful foods (%HFS, healthful food stores) relative to all food stores. Associations between dietary patterns and food source exposure were tested in linear regression models accounting for individual (health and sociodemographic) and area-level (socio-economic and ethnicity) covariates. Montréal metropolitan area, Canada. Urban-dwelling older adults (n 751), aged 68 to 84 years. %FFO was inversely associated with prudent diet (β = -0·105; P < 0·05) and this association remained statistically significant in models accounting for %HFS. %HFS was inversely associated with lower western diet scores (β = -0·124; P < 0·01). This latter association no longer reached significance once models were adjusted for area-level covariates. In Montréal, the food environment is related to the diet of older adults but these links are more complex than straightforward. The absence of significant relationships between healthful food stores and prudent diets, and between fast-food outlets and western diets, deserves further investigation.