PubMed, Cumulative Index of Nursing and Allied Health Literature, Public Affairs Information Service, and Cochrane databases were searched for articles published between January 1, 2002, and June 30, 2013, in English, that reported findings from formative, process, or outcome research on nutrition-related policy and environmental strategies in rural settings. To be comprehensive and capture strategies in addition to those of COCOMO, we searched broadly for nutrition-related policy and environmental strategies applied to obesity prevention. Each search was conducted by using the following terms: rural AND (nutrition or food) AND (community or environment or policy) AND (obesity or overweight or “chronic disease”). In addition to using the search term “rural,” the search was repeated in each database by using predominantly rural states as search terms. The predominantly rural states were identified using the Rural-to-Urban Continuum Codes, the Office of Management and Budget maps, or the Rural Assistance Center’s Frontier map where substantial portions of the state are frontier. The search included relevant references cited in each of the identified studies and in prior reviews of the literature on nutrition-related policy and environmental strategies. NOPREN colleagues also recommended relevant articles.

At least 2 members of the research team screened titles and abstracts and then reviewed the complete text of relevant articles to select articles for inclusion. To be included, the article had to report findings from empirical formative, process, or outcome research related to policy or environmental obesity-prevention strategies in rural communities in the United States or Canada. The term “rural” was broadly defined so as to allow for inclusion of any study in which authors described the setting as “rural,” “non-metro,” “small town,” or “remote” or a study conducted in counties that the Health Resources and Services Administration characterized as rural in 2005 (19). Policy and environmental strategies included, but were not limited to, the 10 nutrition-related COCOMO strategies (Table 1). Although the original COCOMO strategies applied to public service venues, for this study’s purpose COCOMO strategies were expanded to apply to any setting (eg, worksites). Articles that included both rural and urban communities were included only if they reported rural-specific findings.

Data were abstracted from each article by using a standardized form. The form included information about study population (eg, race/ethnicity, socioeconomic status), setting, geographic location, approaches used to adapt the intervention or its implementation to a rural setting, design, methods, and findings. All 17 data abstractors were trained using a strategy similar to that employed by the US Department of Agriculture (USDA) Center for Nutrition Policy and Promotion Nutrition Evidence Library (20). Similar to the USDA’s process, 2 members of the team independently abstracted data, compared abstractions, and then resolved discrepancies by consensus for each article.

Data from the consensus abstraction forms were integrated using data matrices. Four members of the research team reviewed the matrices to identify themes, and tables and narratives were created summarizing data related to those themes.

Of the 29 studies included, 4 took place in Canada (14%) and 25 in the United States (86%) (Table 2). Approximately one-third of the studies (n = 10, 34%) were conducted with American Indian tribes or with First Nations of Canada. The most common settings were schools (n = 13, 45%), small retail food outlets (n = 10, 34%), worksites (n = 5, 17%), and farmers markets (n = 5, 17%). Small retail food outlets were the setting for 70% of studies with American Indian tribes or First Nations (n = 7). About one-third (n = 11, 37%) of the studies took place in multiple settings. Half of the studies (n = 15, 52%) reported findings from formative or process evaluations and did not include outcome data. Fourteen studies (48%) reported findings from an outcome evaluation.

The COCOMO strategy used most often was strategy 1, “increase availability of healthier food and beverage choices” (Table 3). That is, strategy 1 was used in 22 studies (76%), followed by strategy 7, “restrict availability of less healthy foods and beverages” (n = 11, 38%). The strategies used least frequently were strategy 8, “institute smaller portion size options in public service venues” (n = 1, 3%), and strategy 9 “limit advertisements of less healthy foods and beverages” (n = 1, 3%). None of the studies sought to improve the geographic availability of supermarkets (strategy 3).

The literature reviewed yielded 3 themes related to strategy adaptation and implementation in rural communities (Table 3).

Accommodate long distances to food sources when implementing strategies. In 11 studies, investigators discussed how the design and implementation of interventions in rural communities accommodated long distances between food suppliers and retailers and between retailers and consumers. For example, several studies noted that long distances can increase food costs and limit the availability of foods that have limited shelf lives or are sensitive to long transit times (30,39,42). As a result, stores involved in store-based interventions may have trouble stocking the foods promoted through the intervention (40). These challenges are compounded when communities are remote and may rely on specialized transportation, such as annual barge deliveries or food mail programs as seen in a First Nations community in the Canadian Arctic (39). Efforts to address these challenges include strengthening networks between food producers, distributors, and retail food outlets (42). Adaptations also may be required to reduce the distance customers need to travel from their residences to grocery stores and farmers markets (15,23,53) or from worksites to restaurants serving healthy foods or other retail food options (27). For example, farmers markets may increase access by changing the locations where they sell produce (24,37).

Tailor strategies to distinct cultures and food preferences. Investigators of 5 studies identified the need to adapt rural interventions to include specific types of foods. For example, 3 studies reported integrating traditional foods into intervention materials (30,39,48). Another study reported on the importance of understanding cultural values and practices, such as Southern approaches to food preparation (54). The importance of culture was particularly salient in the research conducted with American Indian tribes. For example, 1 study highlighted the importance of engaging tribal leaders, recognizing the history of relationships among tribes, and identifying tribe-specific governance structures, priorities, resources, and champions (29).

Build strong local partnerships when implementing strategies. In 3 studies, investigators noted the importance of partnerships with groups that assist with the redemption of federal food and nutrition assistance program benefits, such as the Agricultural Extension Service (15,24) and Electronic Benefit Transfer administration organizations (28), and parks and recreation departments, hospitals, and health departments (44). Although strong local partnerships are often beneficial in suburban and urban settings, partnerships may be particularly important to leveraging limited resources in rural settings. Also, partnerships may naturally develop in rural communities where social and professional networks are likely to overlap at times because of small populations (55).

Sixteen studies included data on the effectiveness of nutrition-related policy and environmental strategies (Table 4). Most studies (n = 11, 38%) used a quasi-experimental pretest/posttest design with no comparison group. Studies were conducted in 9 settings (communities, health facilities, schools, worksites, faith institutions, farmers markets, small stores, restaurants, and public health departments); some studies occurred in multiple settings.

Twelve of the studies (41%) reporting outcomes documented healthier food environments and policies following the intervention in schools (n = 7, 24%), health facilities (n = 1, 3%), child care centers (n = 1, 3%), restaurants (n = 1, 3%), farmers markets (n = 1, 3%), and worksites (n = 1, 3%).

Ten studies included interventions’ effects on health behaviors or theoretical constructs that are predictive of those behaviors (Table 4). Though results were mixed, interventions tended to improve participants’ intentions to consume healthier foods (34,37,40,47), dietary knowledge (37,47), and self-efficacy related to healthy food acquisition and consumption (40,47). Also, interventions positively influenced the following behaviors: fruit and vegetable purchasing (35), reducing intake of sugar-sweetened beverage (43), and reducing dietary fat intake (47).

Weight status was the only health outcome reported in the reviewed studies (n = 6, 21%) (Table 4). Each of these 6 interventions included multilevel strategies that targeted individual-level behavior change such as counseling and education, in addition to policy and environmental level change strategies that included increasing availability of healthy foods, and discouraging the consumption of sugar-sweetened beverages. Only 1 of the 6 studies reported reducing weight status of participants (36). One study reported that although children’s body mass index increased, the increase was less than in a comparison community (31). Another found that weight status increased (47), and 3 studies found that weight status did not significantly change (34,40,45).

In rural communities, policy and environmental strategies that aim to increase access to healthy foods may also promote economic development through support of farmers, retail stores, and other businesses involved in food production, distribution, and sales (57). Researchers might study strategies that locate retailer’s food outlets in closer proximity to customers, as illustrated by the use of mobile markets by Sharkey et al (58). To tailor interventions to local cultures and taste preferences, those planning rural interventions may benefit from conducting formative work to identify traditional and locally grown foods, as well as local approaches to food preparation. Formative work may also help identify local partners who may be important to promoting and implementing policy and environmental changes in rural areas.

Almost one-third of the studies (n = 10; 34%) were conducted with American Indian tribes or First Nations of Canada. Most of these studies (70%) were conducted in small retail settings (Tables 2 and 3). Research in these often under-studied, at-risk communities is critical to identifying culturally and contextually appropriate approaches to reducing nutrition-related disparities. However, tribally led nutrition-related policy and environmental strategies to prevent obesity may not be generalizable to other rural communities because of tribal governments’ authority to determine their own governance structures, pass laws, and enforce laws through police departments and tribal courts (59). More research can enhance our understanding of the role of tribal self-governance for nutrition-related policy and environmental strategies to prevent obesity (60).

Our aim was to obtain a broad picture of nutrition-related policy and environmental strategies to prevent obesity in rural communities to identify gaps and guide future research. Efforts were made to identify all relevant studies. Formative, process, and outcome evaluation studies were identified for this review, which limited our ability to compare findings across studies, as did what data were collected and reported. Many of the studies were formative. Those studies that assessed outcomes typically involved only a small number of settings and were often quasi-experimental in design. Furthermore, as with all reviews, the study was constrained by limitations in the existing literature and publication bias. Only a limited amount of research on nutrition-related policy and environmental strategies for obesity prevention in rural areas has been published in peer-reviewed journals. The authors recommend consulting websites, gray literature, and other forms of reporting for additional insight into effectiveness and implementation considerations for policy and environmental-level nutrition interventions in rural areas. Finally, we used several strategies to identify studies that were conducted in rural settings; however, studies conducted in rural areas that did not explicitly indicate that they dealt with rural settings may not have been captured in our search.

Explicitly compare the effectiveness of interventions in urban and suburban settings versus rural settings. None of the studies included in the review explicitly compared the effectiveness of policy changes in rural and urban communities. Future investigations should report observed differences in rural settings compared with other settings to inform future research aiming to reduce health disparities in rural areas. Only 14 of the 29 studies identified in this study assessed intervention outcomes at the environmental, policy, or individual level. Therefore, more work is needed to assess policy and environmental, social, psychosocial, behavioral, and biological outcomes associated with nutrition-related policy and environmental strategies.

Experiment with a variety of intervention settings. Among the studies reviewed, the most common settings were schools, small retail food outlets, and worksites. Additional research is needed to explore the feasibility and effectiveness of nutrition-related policy and environmental strategies in other rural settings, such as parks and recreational sites and hospitals, to identify the mix of settings that will yield the greatest population-level reach and effects.

Explore the possibility of aligning federal food and nutrition assistance programs with efforts to increase access to local foods. The limited research to date on COCOMO strategy 5, “improve availability of mechanisms for purchasing foods from farms,” has focused on examining the effectiveness of voucher or coupon programs through USDA. This aligns with a study conducted by the NOPREN Rural Food Access Working Group (RFAWG), examining rural stakeholders’ views about the most promising strategies for improving healthy food access in rural areas, finding that one of the highest ranked policy and research priorities included improving access to federal food and nutrition assistance programs (61).

Report costs associated with implementing intervention strategies. Decision-makers often need information about costs as well as effectiveness when deciding whether to invest in evidence-based nutrition-related policy and environmental strategies (62). Unfortunately, cost and cost effectiveness data are often not reported in scientific articles. In this review, 3 articles included some type of implementation cost information. Conrey et al reported the cost for implementing Women, Infants, and Children (WIC) Farmers’ Market Nutrition Program (FMNP) enhancements across New York State for one year (24); Saksvig et al mentioned that the cost of their school-based intervention was low, but did not provide specific costs (47); and Ruelle et al calculated cost distance, which is a spatial analysis technique that measures costs associated with moving across a landscape to help planners identify potential locations for farmers markets (48). When authors report cost or cost effectiveness information, decision-makers are granted important information from scientific studies that could influence their decision to adopt promising nutrition-related policy and environmental strategies.

Explore the economic impact and the role of local champions related to increasing access to local foods. A recent NOPREN Rural Food Access Working Group study examined rural stakeholders’ views about the most promising strategies for improving healthy food access in rural areas (61). Among the workgroup’s top recommendations was research on the economic impact that strategies have on communities as well as the implications of revenue generation and job creation on increased healthy food access and purchasing power among individuals (61). For example, policy and environmental changes that increase local market and supply chain business opportunities have potential economic benefits for agricultural communities while also increasing access to healthy foods (57). The study’s recommendations align with COCOMO strategies 5 (“improve availability of mechanisms for purchasing food from farms”) and 6 (“provide incentives for the production, distribution, and procurement of foods from local farms”). There is little available research about the effect that local champions, such as policymakers, food policy councils, and other community-driven coalitions, have on nutrition-related policy and environmental change in rural communities. A better understanding could be gained through qualitative work with community stakeholders to determine who local champions are and to identify the best ways to connect with and engage those champions.

These findings help to inform the adaption and implementation of nutrition-related policy and environmental strategies for obesity prevention in rural communities. Although our review was not able to provide policy-makers with information about the effectiveness of different policy approaches, these findings offer insights into the various options available to improve the food environment in rural communities. Moreover, decision-makers should understand the limitations of adopting strategies generated from and tested in geographically diverse settings. The findings also indicate the need for additional research. One major research gap that remains is the limited number of studies testing effectiveness of nutrition-related policy and environmental strategies in rural communities. Future work could identify strategies that have not yet been formally evaluated but that could be feasible in rural communities, such as mobile farmers markets and community garden initiatives.