Duration: 10/01/2006 to 09/30/2011

Administrative Advisor(s):

NIFA Reps:

Non-Technical Summary

Statement of Issues and Justification

Introduction

Numerous concerns over food are reaching epidemic proportions in the US(CDC 2002). The Surgeon General of the United States, for example, describes the problem of obesity as potentially affecting nearly two out of three Americans who are overweight or obese (Carmona 2003). It has been noted by some researchers that increases in obesity rates over the past 15 years have not been accompanied by a related decrease in physical activity suggesting that excessive calories may be the major contributor (Nestle 2003; Putnam et al. 2002). The increased prevalence of obesity and diabetes has occurred at a time when the US food environment has undergone dramatic changes. These changes range from variations in retail food options to shifts in food access and use conditions to consumer perceptions affected by media and cultural values. Consolidation in grocery stores has increased travel distance to stores (and accompanied transportation costs) with implications for an aging population in rural areas and those in poverty. Variations in types and quality of foods available and food prices are reflected in the growth of alternative food sources such as food available at school and work sites (vending machines, fast food places, cafeterias, ethnic and farmer markets, CSAs); changes in personal and community gardening practices; and challenges to safety net organizations such as senior meal sites, food pantries and food banks. Attitudes and values regarding foods also affect the quantity and types of foods consumed. Consumer consumption patterns are influenced by media, local and ethnic cultures, and vested interests in the food production chain. Given all these forms of information and the resources which are mobilized to help shape food choices, there seems to be little evidence of a dominant discourse on food choices. Instead, individuals within the US and many other parts of the world are given numerous options regarding food, and many of these messages are accompanied by statements about good health and/or happiness, regardless of the number of calories consumed.

Some changes in the food environment have increased access to healthier foods. However, many other changes are thought to have increased access and the desirability of high calorie foods and adoption of norms regarding consumption of greater quantities of food. What we do not know is which characteristics of the food environment are detrimental to health and which promote better health. Further, while food systems have undergone global changes, variations among local food environments in prices, access, and retail food options suggest that place matters, not only rural and urban but within neighborhoods and regions. In this project we propose to examine the relationships of food environments to diets, obesity and health. While it is recognized that individuals make food choices, this project assumes that the food environment and foodways are collectively constructed. Thus another aspect of our testing will be the effects of the social organization of food systems including local and regional distribution patterns, alternative food sources, and formal and informal food exchange patterns on diets and health. Our major goal is to determine the impact of food systems on population health including place based populations. Our use of the term "food environment" is a heuristic for our interests in access and use of foods, retail options, and consumer and audience perceptions.

Justification

Many scholars have argued that food is a major component of individual and group identity (e.g., Bell and Valentine 1997; Counihan and Van Esterik 1997; Fischler 1988), as well as the basis for major health concerns (e.g., Ferris 2003; Klaczynski et al. 2004). Given this juxtaposition of a component of culture which is deeply embedded in everyday life for most people, it is important to begin understanding how food patterns are developed. From the perspective that food choices are more than just a matter of taste but involve familiar habits based on socialization (e.g., Bourdieu 1984), agricultural policies (e.g., Goodman and Watts 1997), and availability (e.g., Wrigley et al. 2002), we seek to develop models which highlight the structural and cultural components of the food system and its impact on health.

From everyday experiences, consumers have many choices from which to purchase food. These choices range from internet shopping to large warehouses to neighborhood grocery stores and convenient stores, from roadside stands to farmers' markets to community-supported agriculture, and from fast food restaurants to fine dining. There seems to be ample opportunity to access a range of foods with labels such as organic, fresh, natural, fat-free, and healthy. Access to these various kinds of foods, however, appears to be unequally distributed within the US. One of our goals is to investigate the ways in which access and use conditions within the food environment influence food consumption patterns and obesity rates.

We approach this area of study from a food desert perspective, though we extend our research interests to incorporate a broader scope of food access. Researchers framing food access from the food desert perspective (e.g., Cummins and Mcintyre 2002; Wrigley 2002) seek to measure the density of food stores (and types of food stores) within an area such as a neighborhood, and the average distance someone within that area would need to travel to get to the store. It is often found that in poor areas of cities and rural settings, people must travel further for food, and that the food available may be higher priced, lower in quality and freshness, and less variety. Poverty, obesity and poor health become linked when nutritious foods such as fruits and vegetables are more costly than those high in sugar, fat, and salt. The Food Research and Action Center came to much the same conclusion in their discussions held in November, 2004.

Even when food is readily available, the retail options are often problematic in a number of ways. Schlosser (2001) contends that we have become a fast food nation, consuming hamburgers and fries in record amounts -- two foods which are high in fat and salt. The typical beverages consumed with this meal are carbonated soft drinks or milkshakes, both high in sugar and the latter high in fat. Given our daily caloric needs of 2000-2500 calories (this range varies considerably when taking into account such things as age and level of activity), a meal consisting of a hamburger, fries, and a milkshake accounts for over half those calories (426 calories for a single hamburger, 412 for McDonald's fries, and 720 for an A&W vanilla milkshake -- 1558 total calories).

It is important to understand the food environment in terms of availability of foods such as fruits and vegetables on the one hand, and high fat, salt, and sugar-containing foods on the other. We will gather primary and secondary data in order to assess the availability of foods such as fruits and vegetables and high fat, high sugar foods, taking into account their price, distance from consumers, as well as other difficulties consumers may have in accessing and/or purchasing these foods. And, while there is plenty of information available concerning healthy food choices, Michman and Mazze (1998) have argued that the food industry has become adept at marketing foods which are high in sugar and fat. We propose to investigate further the ways in which retail food stores influence food choices through marketing and promotion. These promotion techniques contribute to how consumers perceive the choices they are given with regards to food.

Based on the work of Hall (1996) and dominant discourses, we seek to map the ways in which food and fitness are discussed in the popular press, experiential knowledge, and community discourse. Within the sociological literature, pathogenic food habits (such as those leading to obesity) are typically framed as a lifestyle outcome (e.g., Smith, Green, and Roberts 2004) or as one end of an eating disorder continuum (with anorexia being on the other) (e.g., Beardsworth and Keil 1997; Klaczynski, Goold, and Mudry 2004). Our study will consist of content analyses of popular media presentations of food choices (and problems associated with food choices) to develop a sense of dominant discources on food, and to help develop survey and interview instruments to measure the ways in which these discourses, if at all, play into values held by individuals with regards to food.

Work on food habits often come to the conclusion that it is the social environment, including the media, educational experiences, and cultural knowledge (or lack thereof), that is important to understand. Much like Gerbner et al.'s (e.g., 1979) work on violence and television, heavy users of television and other forms of mass media are more likely to be obese (Dietz and Gortmaker 2001), as are those of lower educational attainment and few cultural resources. From such a perspective, and coupled with the idea that the US is a mass-mediated society where fast (and fat) food are heavily advertised, one is led to assume that obesity is an unfortunate but unsurprising outcome for many people.

It is hard, though, to simply blame the media given its multivocal nature (Gamson et al. 1992), which seems true concerning food and fitness. Ferris (2003), for example, studied how the media portrayed both Tracie Gold (anorexia) and Carnie Wilson (obesity) as examples of using the star power of these individuals to gain audience attention to the problems of eating disorders. Ten Eyck and Deseran (2004) found that oysters in Louisiana were treated both as a cultural icon and health threat, often in the same issue of local newspapers. A preliminary Lexis-Nexis search on food and fitness shows much the same thing. In August 2005 -- a month chosen at random -- the term "obesity" appeared in 592 articles, while one single food item "ice cream" appeared 691 times. Terms such as "exercise" and "fitness" each appeared more than 4000 times for the same period.

In addition to the media discourse available on these topics, there is also a great deal of public wisdom available, such as cleaning your plate before having dessert, gluttony is a sin, carrying around a few extra pounds can be healthy, heavy people are jolly, sound body -- sound mind, and so forth. We have yet to find an article which says that obesity is appropriate or good, but there seems to be no lack of messages regarding eating (healthy and unhealthy foods), appropriate activities (exercising versus watching television), and normal body types.

At this point it is clear that more work needs to be done on attitudes toward food and fitness, from both individual and societal vantage points. Our approach takes into account culture as argued by DiMaggio (1997) and Swidler (1985), in that knowledge of food and fitness, whether gathered through experiential knowledge, public wisdom, or media discourse (Gamson 1992), are mental schemas/maps which become tools used for navigating through the social landscape. By thinking of this knowledge as cultural tools, and that such tools can be used in a variety of ways, we can make better sense of why individuals who are told one thing by experts think and act in ways which contradict this information.

Related, Current and Previous Work

This section will focus mainly on work that has been carried out by researchers involved with the NC1001 project over the past five years. However, it is important to understand some of the related work that will be used to frame our future research.

As mentioned in the Issues and Justification section, we seek ways to think about food patterns that incorporate social structures in the sense of collective memories or cognitive maps (DiMaggio 1997). Food fits nicely into this way of thinking. It has been found to be strongly linked to identity (e.g., Bell and Valentine 1997; Fischler 1988), so it becomes important to consume specific foods and beverages (and possibly large quantities of these foods and beverages [Harris 1974]) to develop and maintain membership in various groups. For others, rewards come from having a specific body type (muscular, heavy), such as among some African-American communities which view heavy people as those who "have made it" (Witt 1999). Even individuals within countercultural movements will often use food as a source of identity, choosing to purchase and consume foods that are considered grown, processed, and/or retailed in a non-corporate environment (Belasco 1993).

In addition, and as mentioned earlier, we are interested in more structural components such as access. The food desert literature gives us a sense of how easy or difficult it is for people to access healthy, nutritional foods. Together, these two aspects of food and fitness -- access/availability of various kinds of food and the ways in which people perceive food and fitness -- help to position this project within both cultural and structural perspectives. This has been the strength of this project over the past five years.

Our own research over the past five years has included the mapping of food deserts and their role in healthy diets, agricultural systems and local food choices, media portrayals and public perceptions of biotechnology, watershed issues, and the development of food safety messages based on national survey data. Researchers connected to the group have published extensively in peer-reviewed journals, books are in process or have been published, and numerous grants have been awarded to various individuals.

From 2001-2004, members of the research team have received 26 grants valued at over $7 million to examine food, agriculture, community related systems and associations with health. Publications include 37 journal articles, 15 book chapters, three books, 18 technical reports/websites/educational outreach products and over 25 presentations and posters presented. The most significant overall impact of this project is the increased awareness and recognition by policy makers, community leaders, and the public that community organization, the food environment and agricultural practices have significant impacts on human health. A few of the impacts derived from NC1001 research are:

1. Partnerships among public health, community groups, and university researchers in several states (North Dakota, Minnesota, and Iowa) have increased awareness and begun intervention projects that link the community environment and food access research to food insecurity and hunger interventions

*Iowa Department of Public Health has acknowledged the importance of linking the local food environment to food security over the last three years by providing over $160,000 to extend research to local applications. Newly developed educational materials and a hunger website (www.extension.iastate.edu/hunger) in Iowa based on research findings are used by staff and volunteers in the four food banks covering the state and their local food pantry partners to assess food insecurity in their communities and have public dialogues about how to solve local food issues.

*A USDA workshop reaching 160 participants in North Dakota increased awareness of the relationships between food insecurity and the community/local food environment that makes food accessible to rural populations.

2. The public rely on mass media as their primary source of food safety and biotechnology. Food safety issues are often covered in a way that marginalizes the issue and lead the public to see information only when a problem in the food system occurs. Proceedings from a workshop on concerns surrounding future social research on biotechnology held in Baltimore, MD in 2003 have led to meetings in Canada to discuss directions for the Canadian government to fund social research on biotechnology and nanotechnology.

3. Seventy-five educators in New York learned about the impact of food systems on health, and another 22 health professionals have been contacted to think about obesity in New York. In Iowa and New York, 47 community leaders have discussed concerns with food insecurity and are planning the next steps to address these issues.

4. At universities which are represented by members of NC1001, curriculum changes have been put in place to help further develop research and teaching agendas around food issues such as obesity and diabetes.

5. Thirteen apple cider processors in Michigan have been used to investigate the barriers and opportunities to the implementation of HACCP programs within the industry. Over twenty interviews with individuals around the south-central part of the state have been used to educate others on issues such as foodborne pathogens and food handling in a study of communication and food safety. Various other projects have been developed by members to help communities better understand linkages between food and health.

6. Biweekly water quality testing of an impaired Ohio stream at 65 sites over three years reveal high nitrate and fecal coliform levels. As a result the county health department is taking action, three farmer teams are changing farm management practices to lower nitrogen and phosphorous levels in the stream, and one industry NPES permit holder created a nutrient trading program.

7. Residents of a small town in Ohio (pop.1000) now believe that they have excessive nitrate problems as a result of research that showed 23% of the residents have excessive N levels in their drinking water. This knowledge has led to health department interventions and citizens who are ready to implement recommended interventions.

8. Urban gardening offers an alternative resource that provides nutritious foods, urban greening, and sense of community. In NJ, the Healthier New Brunswick 2010 Initiative funded a 2003-4 school-community gardening project recognizing how the activity addresses obesity and chronic disease through exercise and nutrition. At the same time, with Rutgers, it co-sponsored the April 2004 New Brunswick Soil Lead Symposium, responding to findings that very high lead levels have been found in many residential lots, requiring increased testing and risk management strategies.

10. Researchers at University of Minnesota (Dr. Chery Smith), Iowa State University (Dr. Lois W. Morton) and Texas A&AM (Dr. Alex McIntosh) worked together to design a research project related to the food environment and its effects on health and food insecurity.

11. An effort between Iowa and Minnesota uses the concept of food deserts to look at the loss of grocery stores in low-income rural counties and urban neighborhoods and associations with food access, diet, and health patterns. Independently funded (with slightly different goals) by Department of Health in Iowa and the Food Stamp Program in Minnesota, some common methodologies and questions provide opportunities for within state and across state assessments. Researchers are currently in the process of writing two manuscripts that will examine retail food pricing and food security in rural areas, and how civic structure helps solve food insecurity problems for low income, consumers with limited food access in rural food deserts. Each state is also working independently on food systems related projects.

12. Dr. Smith at University of Minnesota has been working with urban homeless families to learn more about how a) they access the food system, b) food systems and social and environmental support mechanisms interact to influence dietary and health quality, and c) homeless people make food choices, when income and access to food is so limited. To accomplish these objectives Smith's team conducted interviews with low-income children, surveyed and measured low-income mothers, and surveyed and measured one child of the mothers. The results from the qualitative data found that temporary shelters and the surrounding community present environments that impact lifestyle choices and ultimately influence the overall health and well-being of homeless families. Participants in the study identified several acute and chronic conditions experienced by their families, including stomachaches, headaches, oral health problems, constipation, psychological stress, weight issues, diabetes, and hypertension. Importantly, participants felt that the shelter environment and policies influenced food choices for their family. Homeless participants in this study expressed the willingness to do anything to ensure their children and families have enough to eat. Key strategies identified by participants included using food stamps; taking food from the shelter facility; stretching food at the end of the month; obtaining food from grocery stores; modifying shopping habits; pawning personal items; sacrificing food for children; and obtaining food from non-traditional sources (dumpster diving).

13. Dr. Morton at Iowa State University expanded her health research beyond civic structure to include the food environment and began to partner with researchers from other disciplines, particularly community nutrition. Her basic research has used the Centers for Disease Control (CDC) mortality data from 1968-1997 to compare age-adjusted US deaths and eight disease specific deaths related to food and diets (diabetes, atherosclerosis, stroke, diseases of the heart, all cancer, breast cancer, digestive cancers, nutritional deficiencies) across the rural-urban continuum based on 1993 ERS urban influence codes. Using five year periods and nine rural-urban categories, she found a persistent lag in lowering mortality rates in rural isolated counties of the US. Further, rural hub counties that have populations between 10,000 and 49,999 not only show lower mortality rates than rural isolated counties but are also lower than suburban counties rates over time. These findings have been published in two places, "Rural Health Policy" Chapter 22 in Challenges for Rural America in the Twenty-First Century edited by Brown & Swanson (2003) and "Spatial Patterns of Rural Mortality" Chapter 4 in Critical Issues in Rural Health edited by Glasgow, Morton, and Johnson (2004). Current and future research using CDC data include the development of statistical multi-level models using county and state level data to examine further how local/state social-economic structure (income inequality, agricultural concentration, occupations, health care institutions, and demographics) are related to age-adjusted county mortality rates.

14. Further in Iowa, a comparison of a purposeful sample of 238 rural and 256 urban low income individuals reveals that 25% rural compared to 9% urban have access to garden produce. Rural low-income individuals were significantly more likely to give food to family, friends, and neighbors and obtain food such as fish, meat, and garden produce from friends and family than urban low-income individuals. There is also a weak but significant relationship between access to garden produce and increased variety of fruits and vegetables in low-income diets. Future research needs to explore this relationship with a more rigorous research design (random urban block sample).

15. Dr.McIntosh (Texas A & M University) has collected data from 300 randomly selected families in the Houston MSA. Data from children in these families include BMI, waist circumference, various skinfold thickness measures, and dietary intake over a three day period. The Texas project also collected a three day activity diary which includes entries regarding consuming food, exercise, sedentary activity, and shopping for food among other activities. Time diaries maintained by the children's parents include entries regarding the amount of time they spent purchasing food, preparing food, and eating food, both at home and away from home. Data analysis has concentrated on the time constraints working parents face and how these affect the scheduling of meals, time spent in family meals, and the effects of these constraints on childrens dietary intake and risk of obesity. Preliminary results suggest that the more time mothers spend with children, the less likely their children are to be obese, but the more time fathers spend with their children, the greater the likelihood their children are to be obese.

Given this level of expertise (which we hope will include people involved in environmental and economic research), and linked to a body of literature on cultural and social structures, we are positioned to shed more light on the ways in which individuals are making food choices, and how local food systems may be developed to help provide for more informed choices.

Objectives

1. Develop models which incorporate access and use conditions within the food environment and how these influence food consumption patterns and obesity rates controlling for individual dietary patterns.
   
2. Study how retail options within the food environment influence food consumption patterns and obesity rates controlling for individual dietary patterns.
   
3. Develop theoretical and empirical models to study consumer perceptions of food within the food environment, and how these perceptions influence food consumption patterns and obesity rates controlling for individual dietary patterns.
   

Methods

Objectives 1 and 2: The first phase of the project will be a community based food system assessment that will characterize variables in the food systems that may relate to high and/or low rates of obesity. The emphasis will be on ranking and scoring potential contributors to obesity while mapping the community's food system. We will conduct participatory rapid appraisal to characterize differences in food systems among communities (see Murray and Scott 1995). A team of researchers will work together in each community under study. In each site, researchers are well acquainted with food-system leaders in their regions. In addition, content analyses of local media and other public arena transcripts will be studied to get a sense of the ways in which public discourse is framing these issues. Finally, we will rely, in part, on secondary data sources for information on the availability of food retailers such as grocers, supercenters, and wholesale food warehouses. These data sources include those maintained by the American Automobile Association (AAA, 2005) and ReferenceUSA database (Version 2002.10) of American businesses, which provide information about sit-down and fast food restaurants (see Neal, 2006). Local.com and ReferenceUSA represent several of the useful sources on grocery stores. These data sources contain information about the size of these operations, as well as their distance from the neighborhoods of our respondents. Given that alternative agrifood institutions are hypothesized to be a key explanatory factor, it is crucial that half of the community study sites have such institutions. This will be determined initially through a review of existing lists of such institutions. Once communities have been tentatively selected, researchers will interview CSA growers, farmers' market managers, and school food-service managers in these regions to verify the presence and vitality of these institutions. Interviews with key informants, including school nutritionists, community health officials and other health professionals who are familiar with community food and nutrition issues, will proceed in all communities. Since this method is participatory, the specific methods used will be co-determined with community members. However, the subtopics in each area that will be explored will include availability of affordable, nutritious food, as well as junk food, alternative food initiatives, civic engagement, local media treatment of food issues, and the types of restaurants and grocery stores in an area. Through this process, we will also build a comparative community data set on all communities to be investigated in the study. This will be a challenge, as not all communities and not all states in the United States collect comparable data on community structure and community health. However, to the extent that it is possible, we will collect information on: 1) the rates of obesity in each community as an indicator of community health; 2) the treatment of sustainable food systems and of obesity in the local media; 3) the level of engagement in community agricultural and food issues by other civic organizations and networks; and 4) the types and density of alternative agrifood institutions. Standard demographic data will also be compiled for each community. This will include information on the ethnic distribution of the population, the percentage of the population living below the poverty level, and the extent of food assistance usage. Discussions of the comparative framework drawn from these incursions will help in thinking about the structural foundations of the local food system. Emergent frameworks will need to be sufficient flexible to provide for changes over the course of the project. This is a grounded theory approach (Corbin and Strauss 1990), which not only offers research flexibility, but has been found to be an effective empowerment tool for local voices. The findings of the community food-system assessment will be reviewed with interested members of the communities. These will be opportunities to identify points of intervention to begin working towards. A community action plan for healthy eating will be an outcome of this process. This type of process has been shown to produce action plans within three to six months, thus being able to take advantage of people's initial interest and commitment. The community will produce a community food-system profile using secondary data. Then, working with local residents and community leaders, the group will conduct inventories and produce GIS maps of the availability and accessibility of fresh fruits and vegetables in the area. At this stage we will identify problems and prioritize areas for action. From this information the group will prepare and produce a graphics-rich publication visually describing the county's food system. We expect to have this aspect of the study completed by November 2008. The second phase of the project will be a quantitative, multilevel survey that will explore both community-level and household-level variables that contribute to obesity, including barriers to healthy eating. The questions on the survey will be informed by the participatory food-system assessments conducted in Phase One in addition to existing research on the factors that relate to obesity with a special emphasis on childhood obesity. The survey will also allow communities participating in the appraisal to compare themselves to other communities. While the design of the survey will be partially dependent on input from key community informants, we envision a questionnaire that would be used to measure a number of key concepts. This would include: 1) the height and weight of family members; 2) the extent to which families purchase or otherwise obtain food that is produced in the local region and distributed through alternative institutions, such as CSAs and roadside stands; 3) Family food goals; 4) Family awareness about barriers to eating healthy in their families and in the community; 5) the relative importance families place on food nutrition vs. cost; 6) Family perceptions about obesity and links to food and eating, and 7) the ways in which large societal values affect food choices. The sample size for the survey will be determined by the population base of each community. The most likely variables hypothesized to be associated with obesity will be operationalized so that they can be obtained through survey questions. Other variables that are known to be associated with obesity from the literature will also be included in the model. The sample will include randomly selected households within the region based on zip codes. To ensure that a sufficient number of families that purchase foods from alternative food institutions are included in the survey, we will target households that purchase food from these institutions. This sample will be augmented by a random sample drawn from the households in the communities, most of whom are assumed not to purchase food from these institutions on a regular basis. We will use multilevel/contextual modeling for our statistical analyses. Suppose we wish to study the relationship between fruit intake and food access. The statistical model would take the form: Yi(fruit intake)=B0+B1(food access)X1+B2(individual)X2+ B3(household)X3 +B4(community/neighborhood dummy variable) X4+B5(interactions between X4 and X1 through X3) Where Yi is a nutrition related outcome such as obesity or food security for a given community, B1 is a vector of food access variables such as the distance to the closest full food grocery store or the number of outlets by type, B2 is a vector of individual characteristics such as age, attitudes, gender; income, education, values; B3 is a vector of household characteristics such as size, availability of functioning transportation (auto, bicycle, bus, etc.), and availability of fruits and vegetables (Bere and Klepp, 2005). We expect to complete this aspect of the project by November 2009. The third phase of the project will involve face-to-face interviews with families selected from the food-system assessments groups and from the survey. The focus of these interviews will be to determine why people do or do not participate in alternative agrifood institutions such as CSAs and farmers' markets, and the effects that participation has on childhood and adolescent obesity. We will also interview families with particularly high and particularly low BMIs to gain in-depth understanding of the choices they make and why they make them. In addition, we will pay particular attention to low-income people, including those using WIC or seniors' farmers' market coupons. Working with the community group, researchers will design questions for face-to-face interviews with families about food systems and obesity. In semi-structured interviews, researchers will ask questions related to food decision-making and obesity, such as food goals, trade-offs, where and why they acquire food, barriers (family and community level) to eating healthy diets, family interaction related to food, focused dietary assessment (e.g. fruits and vegetables), media message awareness and use, level of civic engagement, and weight and height of children and adults in the family. We expect to have this completed by November 2010. In addition to the surveys and interviews, archival research will continue throughout the three phases to follow discourse concerning the local food systems available in various public arenas such as the local media and meetings of political bodies. Content analyses of these documents will be conducted to track trends in reporting, as well as offer further insights into the major actors (or stakeholders) in the area. This content analysis will also be extended to stories generated by local media, so that we can assess the extent to which local media may be promoting civic awareness and the strengthening of local food institutions. The study team will analyze the content to track changes over the three years of the project, in addition to a specific amount of time prior to the beginning of the study to help understand perceptions held prior to the beginning of the project. The project teams in each region will host community meetings in which the findings of the study are presented and discussed. At these meetings, one of the key questions will be how to extend the information generated on problems and solutions related to obesity and the food system. Potential actions could include hosting community meals featuring fresh fruits and vegetables in which the study findings are discussed. The project will also endeavor to develop a core group of community leaders and interviewees who can discuss the issues of food systems and obesity in their local communities and work to effect policy changes. In addition, the project team will work with local newspapers, schools, and Cooperative Extension agents to disseminate the findings of the study. Objective 3: The following is the bases for preliminary work aimed at developing an NIH proposal. We will focus on both state and national news and electronic broadcasting outlets to develop an archive of the available data on food and obesity beginning in 2000. Lexis-Nexis archives many newspapers, including The New York Times and Washington Post, and InfoTrac will be used to search for stories in Time for the same period. Lexis-Nexis only archives three Michigan sources --the Associated Press, Crain's Detroit Business and Crain's Small Business-Detroit. To develop a larger archive of Michigan stories, we will use the Detroit Free Press and Detroit News web pages and microfiche available from the Michigan State University Library. In addition, newspapers which serve the ethnic communities of interest or communities in which large percentages of these groups reside will be contacted to inquire about archival services. The Vanderbilt archive will be used to gather news stories on obesity from ABC and Fox. These networks are chosen based on both ideological leanings and ties to other interests. ABC, for instance, is linked to Disney, which focuses on children, and Fox is considered by some to be the most conservative of the major networks. National Public Radio (NPR) will be used as both a radio source and as a relatively more liberal voice. Television and radio stations in Michigan which serve these ethnic groups will also be contacted, though it is not expected that these stations, especially if they are small, will be archiving their newscasts. If this is the case, they will be recruited to be part of the interviews, as some work has been done on the importance of ethnic news sources (e.g., Olzak and West 1991). Access to information is not the major concern with developing the archive. Instead, it is developing a coding scheme for inclusion and exclusion of stories. Stories not related to food or fitness may have a major impact on eating, such as the death of an important leader (for example, the passing of Yasser Arafat) or family situations such as unemployment or violence. It is possible, though, that discourse on relieving stress may involve eating (Beardsworth and Keil 1997). However, to keep the archive manageable, we must develop criteria for the inclusion or exclusion of articles/stories. For the national level data -- The New York Times, Washington Post, Time, ABC, Fox, and NPR, we will begin with five key words -- obesity (including obese), food, diet, exercise, and fitness. Once articles using these words have been identified, we will further narrow the search by using the following terms -- ethnic, African, black, Hispanic, Latino, Mexican, Middle East, halal, and Arab. Once this second criteria has been applied, we will read/listen/watch the articles/stories to determine its appropriateness for this study. For any year in which there are over 100 articles per ethnic group, a random sample of 100 will be developed. A preliminary search of The New York Times using obesity and Hispanic selected only eleven articles between January 2000 and May 2005 (and no articles for the same time period substituting Latino for Hispanic), so we are assuming that the limit of 100 articles per year per ethnic group will offer a sufficient proxy for the national discourse on this topic. At the local level, the same criteria will be used, though we will not use the second level for newspapers/stations which serve ethnic communities. We assume that if these information sources are running stories on obesity, food, diet, exercise, and fitness, the audience will mainly be the ethnic group. We do not want to limit our investigation of discourse on the subject of obesity to just news stories. As numerous scholars have argued (e.g., Agger 1989; Ritzer 1993; Schiller 1989), information sources include feature stories, advertising, fiction, and so forth. Again, to keep the archive manageable, we will select a limited number of "other" messages which are aimed at getting people to eat and/or exercise. For example, Taco Bell's "I'm Full" campaign would be a good example of a food-related campaign that would fit our study. Three web sites aimed at each of these ethnic communities will be identified and analyzed for information on food and fitness. Finally, pictorial depictions of ethnic individuals will be gathered to offer a sense of what mass audiences are seeing with regards to body shape. We will use ten such depictions for each ethnic group taken from the latest year from when the archive is started. We expect to have this completed by November 2008. Below is a discussion of our approach. These articles/stories will be analyzed using the approach developed by Gamson and Modigliani (1989) in their research on nuclear power, which was further developed by Gaskell and Bauer (2001) for analyzing discourse centered on biotechnology and genetics (Ten Eyck was part of the research team [e.g., Ten Eyck and Williment 2003, 2004]). This approach focuses on how articles are framed, the main actors being quoted or referenced in the articles (which can include institutions), the demands being made, and the overall slant of the article based on a consensus standard. This last part is very important, as coders with various backgrounds can have different opinions on an article. For example, an article which touts the virtues of a surgical procedure for losing weight (such as liposuction) could be seen as negative by someone who has had negative experiences with the procedure and positive by someone who is related to a doctor which performs these kinds of operations. A vegetarian may find an article on an all-protein (aka meat) diet distasteful while someone else may think it is a very good idea, and such concerns with content analyses have been raised by others (e.g., Garfinkel 1967). There are two main foci with regards to the content analyses. First, we are interested in general characteristics of the articles, such as who is considered an expert on topics related to food and health, what demands are being made, how are various fitness and eating strategies framed, and so forth. In addition, we hope to develop a better general understanding of the mass media discourse which has developed (and continues to evolve) on these issues. Blame for the growing obesity problem is often placed -- at least partially -- on the media (or other electronic sources of information such as video games) for advertising unhealthy foods to children or offering an alternative to physical activities. Yet, as has been mentioned, the media are multivalent, and consumers interpret information in various ways. Messages promoting healthy eating and staying fit are readily available. What is unclear is whether some of the trends in the reporting are more salient than others. It is important to develop this area of research to understand why people may be choosing various diet and fitness strategies. Food and fitness are culturally resonate in the US, though other aspects of this topic fall in other realms which may not be considered common knowledge. Discussions of genetics or surgical procedures are not, we assume, a large part of experiential knowledge among the lay public, though national news features of things like liposuction have made more people aware of such procedures. If they know about these things, we expect that they learned it through the media (or from family or friends who read it in the media). For this reason, we will break down the content analyses into subcategories including food (which will encompass stories such as restaurant reviews and recipes), diets and dieting, medical procedures, and fitness. We will only include stories on agriculture or general health if it is food and/or fitness related. Stories on the falling prices of agricultural commodities or the inhumane treatment of food animals, for example, will be ignored. This archive, and its analysis, will provide a way to think about the discourse on food/obesity and health/fitness in a more comprehensive way to deal with the media discourse available than limiting analyses to a few of the major newspapers in the US. (e.g., Gitlin 1980; Ten Eyck and Williment 2003). There will still be gaps in our knowledge, which we hope to partially fill by surveying residents of Michigan, which is the next portion of the section. The survey will be based largely on the content analyses discussed above and on previous opinion/attitude/behavior research done in this area, including a national survey using the internet and a panel of respondents led by Ten Eyck. As mentioned, people get information from a variety of sources, so while the media are considered to be important regarding how people think about food, it is also a topic which many people have a great deal of experiential knowledge. We will ask questions which directly relate to information available through the media (e.g., "Does seeing happy people eating at a restaurant make you want to eat there as well?" "Do you watch food shows or look at the food section in newspapers, and if so, what sort of dishes do you find most appealing?"), in addition to societal values (e.g., "Do you think that Americans should be able to eat whatever they want?" "Do you believe that there should be laws regarding the portions served at restaurants in the US?"), and personal beliefs (e.g. "Are there foods you feel are important to eat to make the right impression on other people?"). As noted in the introduction, the survey must be culturally sensitive. In a study of prostate cancer screening, cultural sensitivity was defined as materials which incorporate ethnic and cultural norms and values (Chan et al. 2003). In this survey, this means asking questions about specific cuisines, role models, and institutions (such as the importance of Ebony for African-Americans or George Lopez for Hispanics). The survey will be developed so that respondents' ethnicity can trigger certain questions and/or answers to get a better understanding of the ways in which culture is linked to attitudes, beliefs, and behaviors. As the survey is being fielded, we will seek out individuals which have been mentioned by the press, through informal contacts, or by asking survey respondents who they feel, if anyone, is a community opinion leader on the topics of food and fitness. The purpose of interviewing these individuals is to contextualize the media messages and survey questions. We assume that some questions regarding food and fitness will be answered by choosing social acceptable answers or a blurring of the "truth" with regards to weight and height (BMIs) and food practices (such as eating out or eating in front of the television). This will give us some idea of the values which are held in the communities, but we need to know if the public wisdom of the community is in line with the survey results. In addition, opinion leaders may be able to offer insight into how to think about changing unhealthy practices within their community. We expect to have the survey completed by November 2009. We will use a semi-formal, open-ended format for these interviews, meaning that we will be focused on specific issues but will ask question that will offer respondents an opportunity to give interpretations of the situation (Denzin and Lincoln 2000). Given the time spent with the archive, content analyses, and survey, we will have some idea of what is happening in each of these communities, but we feel we must avoid constraining respondents' insights into issues of food and fitness in these communities. This approach to interviewing opinion leaders on these issues is original and will help develop a better understanding of what interventions would be acceptable, and how those interventions should be delivered. We expect to have the interviews completed by November 2010. We feel that the above research will be extremely important in gaining a better understanding of how people make choices regarding food and fitness. Without an outreach component, however, we feel the research would simply be an academic exercise. The researchers on this project would like to see the results developed in an applied sense, and we feel it would be most conducive to begin this process by conducting focus groups from members of the various communities under study. The aim of these focus groups would be to share the results from the content analyses, survey, and interviews, and ask what, if anything, can be done to promote a more healthy lifestyle within these communities. According to Morgan (1997) focus groups are best suited for situations in which members would be less inclined to speak to a research in a face-to-face interaction and/or to see how consensus on an issue develops. We expect to have the focus groups completed by July 2011.

Measurement of Progress and Results

Outputs

• As mentioned in the above section on our methodological approaches, we hope to fully engage community leaders and citizens in our research endeavors. Through dissemination of findings in the popular press, interviews with local leaders, and focus groups with citizens, we hope to have a broad impact within communities. Everyone involved in this project has a track record for publications, so dissemination into the scholarly literature is not seen as problematic. Instead, we are most concerned with engagement at the community level.
• Each member state will develop at least one linkage to a minority group within their respective states by extending an invitation to be part of the research program.
• An international conference on local food systems and eating patterns.

Outcomes or Projected Impacts

• Given the uphill battle against obesity regarding the current food situation (agriculture subsidies, millions of dollars spent on food advertising, the location of retail options, societal values on what is appropriate to eat, etc.), we must be realistic in setting goals while staying optimistic. We do believe that we can have an impact at the community level, as well as within the national discourse. One of the major outcomes we seek is to bring the structural and cultural aspects of the food system to the attention of community leaders and reporters. We believe in models such as civic journalism as promoted by Siranni and Friedland (2001), in which antagonists on an issue are brought together to discuss the issues.

Milestones

(2008): A number of milestones have already been reached, including the 26 grants mentioned earlier, and the publications and presentations members of the group have been a part of. As we move forward with this project, we hope to develop stronger linkages with community leaders, and a milestone will be participation in community development which is centered on local food systems.

(2009): Have scholarly and popular publications on the ways in which local food systems impact eating patterns.

(2010): Give presentations to diverse audiences interested in food systems and their effects on food patterns.

(2011): Host an international conference on local food systems and eating patterns.

Projected Participation

View Appendix E: Participation

Outreach Plan

Many of the participants on this project have dual research-extension appointments. Existing and proposed research are strongly linked to stakeholders in each of the participating states ranging from Public health to community leaders working with food security, food banks, and poverty issues to farmers and media. Participatory research in Phase I will feed directly back into the community. Findings from Phases II and III will be shared with surveyed communities and they will be encouraged and supported in responding to our findings. Further, it is expected that many of the relationships begun (reported under impact statements) in the first five years of this project will continue.

In addition, every participant has been active in presenting work focused on food and health at scholarly functions. Some of these functions include the International Food Technologists Annual meetings, the annual meetings of the Rural Sociological Society, and various state and local meetings in which citizens and government officials are present. We will continue to develop the relationships created at these meetings, as well as seek new ties to local and national leaders in these fields.

We expect to have an international conference on food systems and eating patterns in 2011, which will be hosted by one of the member institutions (such as Texas A&M or Michigan State University). This conference will include people from academia, government, the farming community, food pantries, and the food industry. This conference will be used to both disseminate findings from the research and develop stronger ties between these various groups.

Organization/Governance

A Regional Technical Committee will be organized according to procedures in the "Manual for Cooperative Research" dated 1992. The voting membership of the Committee includes one representative from each cooperating experiment station, appointed by the Station Director, and a representative of each cooperating USDA-CSREES research division. Other representatives of participating organizations, including the Administrative Advisor, Administrative Representative, and CSREES representative, are nonvoting members. All voting members of the Technical Committee are eligible for office. The primary role of the Technical Committee will be coordinating the work of the project. Annual meetings will be held for the purpose of conducting business related to the project.
The offices of the Regional Technical Committee will be Chair, Vice-chair, and Secretary. Subcommittees will be named by the Chair. The Chair, in accord with the Administrative Advisor, will notify the Technical Committee of the time and place of annual meetings, prepare agendas and preside at meetings of the Technical and Executive Committees. The Chairperson's responsibilities will include preparation of annual and regional reports. The Vice-chairperson will assist the Chair in all functions. The Secretary will record minutes and perform other duties assigned by the Technical Committee or Administrative Advisor.
The group is currently chaired by Toby A. Ten Eyck from Michigan State University. The Vice Chair is Chery Smith from the University of Minnesota, and Alex McIntosh from Texas A&M is the Secretary.

Literature Cited

AAAnewsroom.net. Restaurant Requirements and Diamond Rating Guidelines. Accessed November 18, 2005.

ACCRA. 2005. ACCRA Cost of Living Index Manual. Arlington VA. Accessed via website November 18, 2005.

Agger, Ben. 1989. Fast Capitalism. Urbana, IL: University of Illinois Press.

Beardsworth, Alan and Teresa Keil. 1997. Sociology on the Menu. New York: Routledge.

Belasco, Warren J. 1993. Appetite for Change (updated version). Ithaca, NY: Cornell University Press.

Bell, David and Gill Valentine. 1997. Consuming Geographies. New York: Routledge.

Bere, Elling and Knut-Inge Klepp. 2005. Changes in Accessibility and Preferences Predict Children's Future Fruit and Vegetable Intake. International Journal of Behavioral Nutrition and Physical Activity 2:15-23.

Bourdieu, Pierre. 1984. Distinction. (translated by Richard Nice). Cambridge, MA: Harvard University Press.

Carmona, Richard H. 2003, September 29. Obesity and Cancer: The scientific links and the future of prevention. Speech given at the National Dialogue of Cancer, Kennebunkport, ME. Available at www.surgeongeneral.gov/news/speeches/obesitycancer09292003.htm

Chan, C.Y. Evelyn, Michelle C. Haynes, Frederick T. O'Donnell, Carolyn Bachino, and Sally W. Vernon. Cultural Sensitivity and Informed Decision Making about Prostate Cancer Screening. Journal of Community Health 28:393405.

Corbin, Julia and Anselm L. Strauss. 1990. Grounded Theory Research: Procedures, canons, and evaluative criteria. Qualitative Sociology 13:3-21.

Counihan, Carol and Penny Van Esterik. Food and Culture. New York: Routledge.

Cummings, Steven and Sally Macintyre. 2002. A Systematic Study of an Urban Foodscape: The price and availability of food in greater Glasgow. Urban Studies 39:2115-2130.

Denzin, Norman K. and Yvonne Lincoln (eds.). 2000. Handbook of Qualitative Sociology (2nd ed.). Thousand Oaks, CA: Sage.

Dietz, William H. and Steven L. Gortmaker. 2001. Preventing Obesity in Children and Adolescents. Annual Review of Public Health 22:337-353.

DiMaggio, Paul. 1997. Culture and Cognition. Annual Review of Sociology 23:263-287.

Ferris, Julie E. 2003. Parallel Discourses and Appropriate Bodies: Media constructions of anorexia and obesity in the cases of Tracey Gold and Carnie Wilson. Journal of Communication Inquiry 27:256-273.

Fischler, Claude. 1988. Food, Self and Identity. Social Science Information 27:275-292.

Gamson, William A. 1992. Talking Politics. New York: Cambridge University Press.

Gamson, William A., David Croteau, William Hoynes, and Theodore Sasson. 1992. Media Images and the Social Construction of Reality. Annual Review of Sociology 18:373-393.

Gamson, William A. and Andre Modigliani. 1989. Media Discourse and Public Opinion on Nuclear Power: A constructionist approach. American Journal of Sociology 95:1-37.

Garfinkel, Harold. 1967. Studies in Ethnomethodology. Malden, MA: Polity.

Gaskell, George and Martin W. Bauer (eds.). 2001. Biotechnology 1996-2000. London, UK: Science Museum.

Gerbner, George, Larry Gross, Nancy Signorelli, and Michael Morgan. 1979. The Demonstration of Power: Violence profile no. 10. Journal of Communication 29:177-196.

Gitlin, Todd. 1980. The Whole World is Watching. Berkeley, CA: University of California Press.

GoRestaurant.net. Your Online Restaurant Directory and Dinning Guide. Accessed November 18, 2005.

Goodman, David and Michael Watts (eds.). 1997. Globalising Food. London, UK: Routledge.

Hall, Stuart. 1996. The Rest and the West: Discourses and power. Pp. 185-227 in Modernity, edited by Stuart Hall, David Held, Don Hubert, and Kenneth Thompson. Cambridge, MA: Blackwell.

Harris, Marvin. 1974. Cows, Pigs, Wars and Witches. New York: Vintage.

Klaczynski, Paul A., Kristen W. Goold, and Jeffrey J. Mudry. 2004. Culture, Obesity Stereotypes, Self-Esteem, and the Thin Ideal: A social identity perspective. Journal of Youth and Adolescence 33:307-317.

Local.com./results.aspx?keyword=restaurants&location. Accessed November 18, 2005.

Michman, Ronald D. and Edward M. Mazze. 1998. The Food Industry Wars. Westport, CT: Quorum.

Morgan, David L. 1997. Focus Groups as Qualitative Research. Thousand Oaks, CA: Sage.

Nestle, Marion. 2003. Safe Food. Berkeley, CA: University of California Press.

Olzak, Susan and Elizabeth West. 1991. Ethnic Conflicts and the Rise and Fall of Ethnic Newspapers. American Sociological Review 56:548-474.

Putnam, Judy, Jane Allshouse, and Linda Scott Kantor. 2002. U.S. Per Capita Food Supply Trends: More Calories, Refined Carbohydrates, and Fats. Food Review. Economic Research Service, USDA

Ritzer, George. 1993. The McDonaldization of Society. Thousand Oaks, CA: Pine Forge.

Rogucka, Elzbieta and Tadeusz Bielicki. 1999. Social Contrasts in the Incidence of Obesity Among Adult Large-City Dwellers in Poland in 1986 and 1996. Journal of Biosocial Science 31:419-432.

Schiller, Herbert I. 1989. Culture, Inc. New York: Oxford University Press.

Schlosser, Eric. 2001. Fast Food Nation. Boston: Houghton Mifflin.

Siranni, Carmen and Lewis Friedland. 2001. Civic Innovation in America. Berkeley, CA: University of California Press.

Smith, Andrew, Ken Green, and Ken Roberts. 2004. Sports Participation and the Obesity Health Crisis: Reflections on the case of young people in England. International Review for the Sociology of Sport 39:457-464.

Swidler, Ann. 1985. Culture in Action: Symbols and strategies. American Sociological Review 51:273-286.

Ten Eyck, Toby A. and Forrest A. Deseran. 2004. Oyster Coverage: Chiastic news as a reflection of local expertise and economic concerns. Sociological Research Online 9,4.

Ten Eyck, Toby A. and Melissa Williment. 2003. The National Media and Things Genetic: Coverage in The New York Times (1971 - 2000) and Washington Post, (1977-2000). Science Communication 25:129-152.

-----. 2004. The More Things Change . . . : Milk pasteurization, food irradiation, and biotechnology in the New York Times. The Social Science Journal 41:29-41.

Witt, Doris. 1999. Black Hunger. New York: Oxford University Press.

Wrigley, Neil. 2002. Food Deserts in British Cities: Policy context and research priorities. Urban Studies 39:2029-2040.

Wrigley, Neil, Daniel Warm, Barrie Margetts, and Amanda Whelan. 2002. Assessing the Impact of Improved Retail Access on Diet in a 'Food Desert': A preliminary report. Urban Studies 39:2061-2082.

Attachments

Land Grant Participating States/Institutions

CA, HI, IA, IL, KS, MI, MN, MO, NC, NJ, NY, OH, OK, TX, VT, WV

Non Land Grant Participating States/Institutions

Cornell University