The physical food environment is hypothesized to play a critical role in influencing dietary patterns, obesity, and diabetes; however research on this topic has been largely cross-sectional. Consequently, little is known about the long-term associations between access to away-from-home food (restaurants and fast food places), patterns of away-from-home eating, and health outcomes. Objectives: This study seeks to examine longitudinal changes in the physical food environment and determine their effect on diet and health outcomes. Specifically, we hope to (1) determine factors in the built environment that are associated with the number of away-from-home food places in an individual's residential neighborhood; (2) determine the relationship between the number of away-from-home eating places in an individual's residential neighborhood and frequency of away-from-home food consumption; and (3) determine the longitudinal associations between the frequency of away-from-home eating with 13-year changes in weight and glucose status. Methods: Data collected from the Coronary Artery Risk Development in Young Adults (CARDIA) Study, a longitudinal cohort of 5,115 young adults, will be used in these analyses. Individual-level health and diet data have been linked by residential location to a Geographic Information Systems (GIS)-derived database which includes measures of the food environment (i.e. location of restaurants), environmental features (i.e. road networks), and census measures (i.e. block-group average household income), resulting in a complex database of both individual and environmental factors. We will use a combination of multivariate, hierarchical linear regression ([HLM] a form of multilevel modeling), and longitudinal fixed-effect models. HLM allows us to examine the independent contribution of both individual and environmental factors to the frequency of away-from-home eating. Fixed effect models, used to examine inter-individual changes in behaviors and health outcomes over time, have the additional benefit of providing inherent control for endogeneity (unobserved, time-invariant heterogeneity within individuals). Implications: Weight loss and weight gain prevention efforts focused solely on individual-level factors have been met with limited success. Understanding influences of the community food environment on dietary choice and health-related outcomes may be necessary for promoting healthy lifestlyes. Results from this study will identify factors that could be targeted for the design and development of community environments that improve individuals' diets and support healthy behaviors. ? ? ?
| Duffey, Kiyah J; Gordon-Larsen, Penny; Shikany, James M et al. (2010) Food price and diet and health outcomes: 20 years of the CARDIA Study. Arch Intern Med 170:420-6 |
| Duffey, Kiyah J; Gordon-Larsen, Penny; Steffen, Lyn M et al. (2009) Regular consumption from fast food establishments relative to other restaurants is differentially associated with metabolic outcomes in young adults. J Nutr 139:2113-8 |
