Even among a sample of largely chain supermarkets, availability of healthier foods may present an obstacle to consuming a healthy diet in certain neighborhoods, particularly low-income neighborhoods, consistent with findings from previous research [8
]. Further, there appears to be a significant disadvantage in the availability of low fat milk in predominantly African-American neighborhoods. However, when available, the quality of healthier food items did not differ in supermarkets and grocery stores based on the characteristics of the residents of the store neighborhood. Prices for healthier food options also did not vary markedly in the different neighborhoods studied, although larger stores consistently offered more favorable prices for healthier options, as has been shown in previous studies [6
]. These findings are of particular concern as areas that are more vulnerable to diet-related chronic diseases (i.e., lower income, rural, and predominately African American or Hispanic) typically have limited access to large food stores [4
Furthermore, we found that the overall price for the 10 non-produce healthier items on the measure was significantly greater than the price for the standard items, which was primarily driven by the higher prices for healthier juice, hot dogs, ground beef, chips and whole wheat bread. Given that these foods are frequently consumed in the typical American diet [8
], there may be a significant disincentive, in general, for purchasing the healthier version of these common food items. Moreover, the use of brand name products as required by the NEMS-S protocol may actually underestimate the price differential between healthy and regular products, as individuals who are working within a very limited budget are unlikely to choose brand name products and might instead choose the store brand products or the products that are available in bulk sizes (which may not be the healthier version of the product). Examining the store brand and bulk products may produce even greater price differences between healthier and regular food items.
In addition to general healthy food availability differences based on neighborhood median household income and trends related to neighborhood racial composition, there may be some specific disadvantages in the amount of low-fat milk available in certain neighborhoods. Consistent with previous research, availability in terms of low-fat milk shelf space was significantly lower in store neighborhoods with lower median household income [7
] and with higher proportion of African-Americans [5
]. However, a casual relationship cannot be determined from these data; supermarket availability of low-fat milk may lead to lower demand for this product among certain populations, or lower demand for low-fat milk may lead to lower availability of this product in supermarkets. Nonetheless, lower shelf space for low-fat milk in vulnerable neighborhoods may create lower familiarity with low-fat milk or may convey social norms for drinking higher fat milk. Lack of availability of lower-fat milk in populations at high risk for chronic disease may be a crucial public health concern as whole milk is one of the top dietary sources of saturated and total fat for Americans [29
] and switching to lower fat milk options is a priority in several dietary improvement interventions such as the 1% or Less campaign [30
] and the CDC’s Rethink Your Drink campaign [31
When interpreting the findings of the current study, it is important to consider several limitations. First, although we followed the conventions of previous research and utilized median household income as a proxy measure of the neighborhood socioeconomic status [7
], other researchers have selected other data to characterize socioeconomic status of neighborhoods, such as median house value [25
] or poverty rates [4
]. We found comparable results when using poverty rates rather than median household income and therefore have some confidence that socioeconomic status is reasonably estimated in the current study. However, future research may wish to focus on determining the most appropriate measures for socioeconomic status when characterizing the food environment. Second, previous studies have used a variety of definitions for the space in which people live (such as zip code, census tract, or state) and census tract may not be the most appropriate way in which to characterize grocery store and supermarket neighborhood; however, the most appropriate unit of analysis in evaluating neighborhood food environment has yet to be determined [34
]. Finally, a convenience sample of grocery stores and supermarkets selected by overweight individuals entering a weight loss program was used. The extent to which this sample generalizes to the primary food stores chosen by other populations, such as non-overweight individuals or overweight individuals who are not entering a weight loss program, is unknown. The sample of food stores assessed represents a third of the available stores within the defined catchment area, but the possibility exists that these stores are not representative of the larger group of grocery stores and supermarkets. Because overweight and obese individuals are a group at elevated risk for developing the chronic diseases associated with energy imbalance [35
], understanding the neighborhood food environment of the grocery stores in which this subset of the population shops may be of particular importance. Furthermore, Arkansas and Vermont are rural states and even the large cities within these states are not as densely populated as others have studied [6
]; therefore, generalization to large metropolitan areas with significant inner city “food deserts” may be problematic. However, rates of obesity and diet-related chronic disease within rural states are alarming [36
], and these regions should not be ignored when considering the built food environment. Indeed, research shedding light on the food environment within rural states provides a necessary complement to investigations of urban metropolitan settings and rounds out the existing scientific literature to provide a more complete picture of the nation’s food environment.
Despite these limitations, these results point to possible targets for improving the accessibility of healthy foods. Environmental and policy initiatives can improve conditions for large numbers of individuals and the growing concern with the impact of the food environment on the health of the nation emphasizes the importance of considering such actions [38
]. Given these current results that healthier foods tend to have more favorable prices in larger stores and lower-income areas may have reduced availability of healthier items, efforts to increase supermarkets accessibility may be particularly beneficial in developing food environments that support the prevention of diet-related conditions, especially among lower income populations and other high risk populations. Cost is reported to be only second to taste in determining food choice [39
], and price reductions for lower fat foods as well as fruits and vegetables have been found to be effective in increasing healthier food purchases [40
]. However, supermarket operators are likely reluctant to locate in certain areas due to crime/security, sales projections, development costs, and insurance costs [17
], which creates a challenge of healthy food accessibility in these areas.
There have been recent movements to expand the focus on traditional individually-targeted interventions to reduce nutritionally-related conditions such as obesity, heart disease and cancer to also include a focus on modifying the built environment to impact these nutrition–related public health challenges. In one such case, the Food Trust is taking an innovative approach to improve the accessibility of nutritious food for underserved urban communities in Pennsylvania by assisting with the financing needs of supermarket/grocery store operators that plan to establish a store in underserved communities. So far the Food Trust’s Fresh Food Financing Initiative has committed resources to over 50 supermarkets projects [41
]. While the efficacy of neighborhood-level food environment interventions to promote healthier intake remains to be seen, it is possible that economic incentives or community advocacy for locating supermarkets in underserved areas could significantly improve the food environment for populations at-risk for nutrition-related chronic conditions.