Background

The goal of this systematic review was to determine whether food access (i.e., the availability and affordability of foods) is associated with weight status. Food availability was measured in many ways for individuals, families and neighborhoods, including proximity to food outlets, density of food outlets, distance to the outlet closest to a residential location, mean distance traveled to food outlets where participants reported shopping and foods available in stores. Food availability was measured both objectively (e.g., food outlets per capita or within a certain radius of a participant’s home using geographic information systems) and subjectively (e.g., perceived availability of healthy vs. unhealthy food outlets and healthy foods in a store or neighborhood).

Conclusion statement

Limited but consistent evidence suggests that the relationship between access to convenience stores and weight status is unfavorable with closer proximity and greater access being associated with significantly higher body mass index (BMI) and increased odds of overweight and obesity.

2015 DGAC Grade: Limited

The body of evidence on access to other food outlets, such as supermarkets, grocery stores, farmers’ markets and produce stands, and weight status is limited and inconsistent.

2015 DGAC Grade: Not assignable

Methods

Literature searches were conducted using PubMed, Embase and Cochrane databases to identify studies that evaluated the association between food access and dietary outcomes. Studies that met the following criteria were included in the review: Randomized controlled trials (RCTs), non-randomized controlled trials, prospective cohort studies, or cross-sectional studies; humans over the age of two years who were healthy or at elevated chronic disease risk; subjects from the United States (US); and published in English in peer-reviewed journals. The date range was January 2004 to January 2014. The intervention or exposure was food availability and accessibility, compared to different levels of availability and accessibility. The outcomes were dietary intake, quality, behaviors and preferences. Data from each included article was extracted and risk of bias was assessed. The evidence was qualitatively synthesized, a conclusion statement was developed and the strength of the evidence (grade) was assessed using pre-established criteria including evaluation of the quality and risk of bias, quantity, consistency, magnitude of effect and generalizability of available evidence.

Findings

• This NEL systematic review included 26 articles published since January 2004 that examined the relationship between food access and dietary outcomes. This literature included 19 cross-sectional studies and seven longitudinal studies
• The studies used multiple approaches to assess food access (i.e., the availability and affordability of foods) and dietary intake, quality and variety. The majority of studies measured food access by the density of food outlets within a specified distance from a participant’s residence and proximity to various food outlets. The primary weight status outcome of interest was BMI.
  • Although food access was assessed across geographic, ethnic, racial and income groups, making comparisons across studies was challenging because of the wide variation in methods used to determine food access. Despite this variability, the relationship between convenience stores and weight status was consistent across the evidence.
  • Seven studies found statistically significant associations between access to convenience stores and BMI, as well as increased odds of overweight and obesity. Five of these studies were completed in an adult sample; two assessed this relationship among children.
  • Due to the variability of studies and paucity of data, no consistent associations regarding dietary outcomes and access to other food outlets were evident.

Limitations

The ability to draw strong conclusions was limited by the following issues:

• Most studies were of weaker design (i.e., cross-sectional)
• The methodological differences across individual studies were significant
• The myriad retail outlets that sell food vary considerably.