Food Acquisition and Shopping Patterns in the United States: Characteristics and Relation to Body Mass Index in the US Food Acquisition and Purchase Survey

• Published in: Journal of the Academy of Nutrition and Dietetics 2022-04, Vol.122 (4), p.745-757.e2
• Main Authors: Ma, Xiaonan, Bell, Bethany A., White, Kellee, Liu, Jihong, Liese, Angela D.
• Format: Article
• Language: English
• Subjects: Body Mass Index; Commerce; Cross-Sectional Studies; Diet; Farmers’ market; Food bank/pantry; Food Supply; Grocery store type; Humans; Shopping distance; Transportation; United States
• ISSN: 2212-2672; 2212-2680
• DOI: 10.1016/j.jand.2021.09.013

Previous studies examined the association between shopping distance, frequency, and store type separately. The objective is to explore food acquisition and shopping habits using multidimensional measures and examine its association with body mass index (BMI). A cross-sectional study was conducted. Four thousand four hundred sixty-six households from the US Food Acquisition and Purchase Survey during April 2012 to January 2013 were included in this analysis. Both continuous BMI and categorical BMI were used. Latent class analysis was used to identify the latent profiles using travel distance and perceived travel time between residential location and primary store, store type, transportation mode, and farmers’ market utilization. Multivariable linear regression and multinomial logistic regression were used to assess the association between the identified patterns and continuous and categorical BMI. All analyses were stratified by urbanicity. Overall, 65% (weighted percentage) of households were located in an urban tract. Thirty-seven percent were categorized as Class 1 (households that shopped more proximally, used their own vehicle, and shopped at a farmers’ market), 50% as Class 2 (households that shopped more distally, used their own vehicle, and shopped at a farmers’ market), and 14% as Class 3 (households that shopped proximally but perceived longer travel time, used someone else’s vehicle, and did not shop at a farmers’ market). Among rural households, 54% were Class 1 and 46% were Class 2 (Class 3 was not identified). Socioeconomic status characteristics, proximity, and store food price concerns were associated with the identified patterns. However, no significant association was found between the identified patterns and BMI. Food acquisition and shopping patterns were not associated with BMI in this national sample. However, future studies should also investigate the role of economic factors, such as food prices, in relation to shopping patterns and BMI.