The Public Health Effects of Food Deserts: Workshop Summary (2009)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

6 Research Gaps and Needs As described throughout the two days of the workshop, the purpose of food desert research is to understand factors that contribute to food deserts and ultimately to identify ways that facilitate change for health and non-health benefits. This is an emerging field that brings together a variety of disciplines, including public health, nutrition, economics, geography, and urban planning. The final session of the workshop sum- marized how additional research is essential for clarifying the causal link between the food environment and health and for informing researchers when they develop the most promising interventions. The study of food deserts and determining their impact on public health is extremely com- plex and requires multidisciplinary research approaches. An overriding message of the workshop was that evidence shows food deserts exist in the United States along income, ethnic, and racial lines, both in urban and in rural areas. There are rich data on the local level and more general information on the national level, but it is yet to be determined if these findings from one area can be broadly applicable to areas with similar demographics. The food environment is dynamic. New players are taking a growing percentage of the consumer’s food dollar; these include supercenters, in particular, as well as farmers markets, dollar stores, convenience stores, and other outlets. Focusing only on supermarkets and grocery stores ignores the places where millions of Americans purchase some or all of their food. 67 

68 THE PUBLIC HEALTH EFFECTS OF FOOD DESERTS Below are topics that were raised during the panels and summarized in this final wrap-up session. DEVELOPMENT OF METHODOLOGY AND TOOLS “Food access” was defined by presenters in various ways. A standard definition could help in making appropriate comparisons and furthering insight. One challenge is to resolve how the definition should incorpo- rate such factors as geography, economics, and choice. If food access is determined using a spatial scale, the definition of “neighborhood” would benefit from further clarity and refinement. Researchers reported on complementary instruments to measure food availability and affordability, including GIS, market basket surveys, other survey instruments, census data, and the Consumer Price Index. There is a strong desire (1) to develop and/or refine rigorous measures that are sensitive to the needs of diverse populations, and (2) to incorporate qualitative methods into research, in order to provide better information about issues such as consumer perceptions of food access. Currently available data about food outlets from both public and pri- vate sources lack validation. Several presenters uncovered errors within national databases (such as those from Dun & Bradstreet and infoUSA) when they validated the measurement of these spatial data sets against actual visual measurements in specific neighborhoods, but it is unknown whether errors are random or somehow skewed to bias results. In addi- tion, new census data, when available, will need to be used to investigate whether changes in the number of food outlets reflect population shifts. As noted above, the generalizability of local studies needs to be known before interventions can be applied on a broad scale. Epidemiological Methods Combined with Multidisciplinary Approaches Qualitative methods are also important for understanding the nuanced interaction between personal preferences and perceived access to quality food, which can then be compared to what is actually available. Challenges to understanding the links between food and health remain; they may best be met with multiple types of evidence from rigorous observational studies, natural experiments, simulations, and evaluations of evidence-based actions. Methods and tools from geography, demogra- phy, economics, psychology, sociology, urban planning, and policy can all help inform epidemiological research. 

RESEARCH GAPS AND NEEDS 69 Longitudinal Studies To date, most studies have been cross-sectional in that they compare different areas with different food environments. Longitudinal studies are crucial because they provide valuable information as the research follows the same population over time. Natural experiments can provide good information, but it is important that the experiments are theory driven and nuanced by population and other variables. Longer time frames are often critical for judging the effect of different interventions and possibly linking a population’s food environment to its health. For natural exper- iments and other interventions, better surveillance methods can help researchers track information to see how an area is changing over time. Policy and Program Evaluation Policy and program interventions—such as those described in ses- sion five of the workshop (see Chapter 5)—were not generally set up by researchers. However, these activities may provide important opportu- nities for evaluation. To learn from both the successful and the unsuc- cessful elements, researchers could set up benchmarks for performance, sampling strategies, pre-testing of instruments, measurement of impacts on different sociodemographic groups, and process evaluations during interventions. APPROACHES TO MEASURING FOOD DESERTS AND OUTCOMES A theme that ran throughout the workshop was recognition of the complex physical and social environments in which food deserts are located. Approaches to understand some of the barriers can come from different disciplines working together. Epidemiological The causal links between food deserts and health have not been firmly established. Researchers may need to look at more proximal behav- ior changes, such as shopping behavior, and then look at dietary behavior and ultimately disease outcomes and weight. Understanding the link between food availability and changes in obesity requires a better under- standing of these intermediate steps, particularly the effect on dietary intake and shopping and eating behaviors. Research shows that people do not adjust caloric intake when they consume calories via beverages. It is not understood why this is so, nor 

70 THE PUBLIC HEALTH EFFECTS OF FOOD DESERTS are the implications clear for dietary recommendations. Randomized con- trolled trials on the effects of low-fat versus whole milk could be con- ducted, especially because children over age 2 and adults are currently advised to drink lower-fat milk. Individual foods, overall diet quality, dietary patterns, and meal size and frequency play different roles in health outcomes. Because it is impor- tant to select the right interventions on which to focus resources, it makes sense to understand better which ones make the most difference. Geospatial and Demographic Although researchers aggregate and analyze data by geography, peo- ple may define their neighborhoods differently, in both urban and rural settings, which makes a difference in how they define food access. Their definitions may or may not coincide with administrative boundaries, cen- sus tracts, or other top-down categories. A better understanding of spatial behavior moves from merely the supply of stores, or of food within stores, to how people make decisions based on the spatial features around them. In addition, research to date has used the home as the central point, yet some people shop after work or in combination with other places they frequent during their day. Formative research, with participation by community members, can help explore what stores can reasonably offer as healthier options that satisfy the preferences of consumers from diverse cultures. An example is the food dynamic in Latino communities, in terms of whether tiendas mitigate against food deserts more than small stores in other communi- ties, as well as the relatively high use of whole milk. Different issues across the lifespan came to light during the work- shop. These issues include: • the connection between food deserts and child BMI; •  eenagers’ sensitivity to price, marketing messages, and eating t p ­ atterns that fit their schedules; •  arents’ purchasing decisions for themselves and their families; and p •  ging adults with limited mobility, especially in areas without a public transport. Economic The presentations underscored consumers’ responsiveness to price. Additional research is necessary to gauge the impact of changing the prices of healthy and less healthy foods, as well as how purchasing behav- iors vary by income, age, racial, and ethnic group. Will there be improved 

RESEARCH GAPS AND NEEDS 71 food and beverage choices if we increase the prices of caloric bever- ages, whole milk, or other unhealthy food choices or conversely reduce the prices of water, diet beverages, and low-fat milk? Research on price manipulation within stores as a public health intervention is limited, in part because of stores’ reluctance to share pricing data. Urban regenera- tion stores in the UK may be more willing partners in this research than other commercial enterprises. In addition, many argue that it is important to evaluate how individu- als’ perceptions affect their food access. People may have physical access, but not perceive they have economic access to healthy food. Endogeneity is another issue that could benefit from further explora- tion. An endogenous factor or variable is one whose outcome is predicted by many of the same factors that arise within the model being studied. There are usually many unmeasured factors that affect both the endog- enous factors and the outcome. For instance, access is endogenous to food deserts and health outcomes: those with access to supermarkets may have better diets because they choose to live near supermarkets, as healthy eating and nutrition are part of the decision-making for where residents locate. Alternatively, another endogenous factor might be the existence of a genetic susceptibility that enhances the taste of fat, increasing intake of fried fast foods and also affecting weight gain. Endogeneity may explain a great deal of the cross-sectional associations between measures of the food environment inside food deserts and food choices, obesity, and other health outcomes. As presented at the workshop, small stores are abundant in urban and rural areas, yet usually carry little healthy food. The cost of purchasing new refrigerators and sinks in which to prepare and store perishable items is often prohibitive for many small store owners. Fear of unsold or spoiled stock may also contribute to their reluctance to purchase perishable food inventory. Finding ways around these barriers could benefit both consum- ers and storeowners. Finally, the effect of the recession on consumer food choice and store survival may require more clarification, because the state of the economy is uncertain. Development of a method to assess total cost, to include price, food access, preparation time, and convenience, would assist in understanding the situation in a particular community and ways to improve it. Social Sciences The social, cultural, and psychological factors that influence human behavior are clearly relevant to explain how consumers interact with the food environment. The most effective interventions will be those that are sensitive to the needs of diverse populations. 

72 THE PUBLIC HEALTH EFFECTS OF FOOD DESERTS On the supply side, in addition to the economic issues, there may also be a psychological component to why supermarket executives are reluctant to site stores in food deserts and why a store does not want to be the first of its kind in a neighborhood. This is in contrast to their UK counterparts that seem to be more willing partners in regenerating urban stores. If data can back up this perception, perhaps more targeted policies can ensue. Psychosocial factors that affect small-store owners and shoppers are also important to explore, such as feelings of self-esteem, stress, and locus of control. As examples, these factors may affect how store owners and employees interact with customers, as well as how willing customers are to purchase and prepare healthy foods that might be new to them. The roles that food venues play in the community and for individuals go beyond places for commercial transactions. Urban Planning The public health implications of zoning and transportation are addi- tional areas in which urban planners can contribute their expertise. The lack of transportation to existing stores is an issue for some because people do not have the means to travel outside food deserts, yet the popu- lation may be too small and dispersed to support new markets. This is an area of study where urban planners could help develop useful approaches to improving transportation infrastructure so that those lacking private transportation could gain access. The current planning paradigm favors mixed-use “smart growth,” an urban planning concept that clusters growth in the center of a city to create more walkable, lively neighborhoods and urban areas. Still to be understood is the role the food environment plays in where people choose to live and how smart growth affects health. Policy Policy makers could use the available findings to develop policies at federal, state, and local levels that are intended to improve dietary behav- ior while recognizing that many unknowns may affect final outcomes. Some of the policies identified in the workshop that will benefit from filling research gaps include the following: •  ow federal and other government benefit programs can encour- H age healthier eating; •  he effect of taxing unhealthy food (especially sweetened bever- T ages) and/or subsidizing healthy food; and 

RESEARCH GAPS AND NEEDS 73 •  he best mix of financial and other incentives to site supermarkets T in food deserts and to encourage existing stores to stock healthy items. In concert with additional research, it is important to connect concep- tual data with the context of people’s everyday lives and to fully explore the consequences, often unintended, of decisions around such a perva- sive, personal issue as food choice. NEXT STEPS AND CLOSING THOUGHTS Improvements in the methodology of food desert research will be helpful for developing evidence-based, locally appropriate interventions. The interplay between supply and demand is complex. Many workshop participants expressed that a supply of healthy food needs to be avail- able and affordable for consumers to purchase and prepare on a regular basis. However, focusing only on supply, especially when healthy items cost relatively more than less healthy options, may not have a significant impact on the health of individuals or broader communities; consumer demand—in the forms of preferences and knowledge—also affects con- sumption decisions and subsequently health outcomes. In addition, a caveat ran throughout the workshop that retail is but one part of a larger food environment in which both healthy and less healthy choices abound. Consumers get food messages from sources that range from the media, to family and friends, schools and other educational outlets, and underlying cultural norms. Businesses operate on thin profit margins and they con- stantly balance customer demand for both healthy and less healthy (but often good-tasting) choices. To close the workshop, Barry Popkin, planning committee chair, thanked staff, speakers, and participants. Understanding food deserts is the beginning of a long set of issues to understand how to improve the diets of Americans.