Article Preview
TopIntroduction
According to recently released US dietary guidelines, recommendations for adults include 2.5 cups of vegetables and 2 cups of fruits per day (USDA, 2010). However, the average American’s consumption of fresh produce is far below recommended levels, and only 6 percent and 8 percent of individuals achieve this recommended target for vegetables and fruits, respectively (NFVA, 2010). Many studies have found that increased consumption of fresh fruits and vegetables reduces the risk of cardiovascular diseases, type 2 diabetes, obesity and certain types of cancer (He et al., 2004; Hung et al., 2004; Johnsen, 2004; Villegas, Salim, Flynn, & Perry, 2004; Willett, 2005).
A growing number of studies have shown that adequate access to healthy foods leads to increased fresh produce consumption and reduced risk of chronic diseases (Bodor, Rose, Farley, Swalm, & Scott, 2008; Lane et al., 2008; Larson, Story, & Nelson, 2009; Rose & Richards, 2004; Zenk, Schulz, Hollis-Neely, et al., 2005). The concept of food access has multiple meanings – from spatial accessibility to the nearest food vendor to price and variety of food items in the store. Among the various food access studies, more research has been done on spatial accessibility because it can be measured or modeled in Geographic Information Systems (GIS) and does not require extensive field work.
There are two main groups of spatial accessibility measures – density-based and distance-based. Density-based measures represent the number of food vendors per person or per area unit within a geographical area, often delineated as an administrative and enumeration unit (e.g., census block, census tract, zip code area, or municipality). The main disadvantage of this approach is the assumption that residents of a given unit shop in stores located within this unit. Distance based measures range from a simple linear distance “as the crow flies” to a food vendor (Michimi & Wimberly, 2010; Winkler, Turrell, & Patterson, 2006; Zenk, Schulz, Israel, et al., 2005) or a distance based on road networks (Algert, Agrawal, & Lewis, 2006; Goldsberry, Duvall, Howard, & Stevens, 2010; Smith et al., 2010) to cost distance that allows the inclusion of slope, land use and physical barriers such as railways and rivers in the analysis (Burns & Inglis, 2007).
Multiple studies have documented lower spatial accessibility of fresh produce in more disadvantaged neighborhoods in the US and Australia (Burns & Inglis, 2007; Morland, Wing, Diez Roux, & Poole, 2002; Powell, Slater, Mirtcheva, Bao, & Chaloupka, 2007; Zenk, Schulz, Israel, et al., 2005). These studies use census data on income, median house values and racial composition (Morland, et al., 2002; Powell, et al., 2007; Zenk et al., 2006), or racial segregation or a composite disadvantage/deprivation index to identify and characterize neighborhoods as disadvantaged.
However, there is also a growing body of literature suggesting that the most disadvantaged neighborhoods have better spatial access to supermarkets and other sources of fruits and vegetables than less disadvantaged areas in the US, UK, Canada (Apparicio, Cloutier, & Shearmur, 2007; Macdonald, Ellaway, & Macintyre, 2009; Pearce, Hiscock, Blakely, & Witten, 2008; J. Sharkey, Horel, & Dean, 2010; J. R. Sharkey & Horel, 2008; Smith, et al., 2010; Smoyer-Tomic, Spence, & Amrhein, 2006).The disagreement in findings could be due to the fact that most of these studies are conducted at a local scale (one or several cities or communities) and thus do not allow for a comparative analysis and identification of more general trends at a larger spatial scale.