Abstract
Background Understanding the most important elements of nutrition environments that affect eating behaviors can inform public health nutrition policy and programs Conceptual models depict interrelationships between these elements. However, empirical testing of conceptual models of nutrition environments and eating behaviors has been limited. The purpose of this study was to explore which factors in nutrition environments are significantly associated with dietary outcomes in two urban, low-income, and minority food desert communities.Methods This study analyzed cross-sectional baseline data from the Food in Our Neighborhood Study (FIONS). FIONS data were from adults recruited based on a random sample of addresses in neighborhood study areas of Philadelphia, PA and Trenton, NJ that were each three square miles and designated as low supermarket access areas. Study participants were required to live within one of the study areas and be the primary adult food shopper for the household. Study participants responded to a survey with ten domains that included shopping preferences, grocery spending, home food availability, perceived neighborhood nutrition environment, background characteristics, and ASA24® dietary assessments. Store audits were conducted in both study areas to estimate observed nutrition environments. Exploratory factor analysis (EFA) was conducted on over 120 survey items and yielded a model of four factors with 22 items. MIMIC model analyses were performed on these four factors controlling for covariates. Main dietary outcomes were Healthy Eating Index (HEI) scores and fruit and vegetable consumption subscores computed from ASA24® assessments.Results The sample included 796 participants (n = 393, Philadelphia, PA; n = 403, Trenton, NJ) with demographics representative of urban, minority and low-income food desert communities: 60% were African-American, 55% had annual household incomes <$30,000, 45% participated in SNAP or WIC, and 58% experienced at least some food insecurity. Among four factors that emerged in EFA, three (My Store’s Quality; Perceptions of Neighborhood Food Availability; and Household Food Challenges) were significantly correlated with vegetable consumption subscores, and one (Household Food Challenges) was significantly associated with HEI scores.Conclusions This research tested and confirmed the importance of perceived nutrition environments and household food challenges in predicting dietary outcomes among residents of two urban, low-income, and minority food desert communities.