key: cord-351785-d35kqobp authors: DeWitt, Emily; Gillespie, Rachel; Norman-Burgdolf, Heather; Cardarelli, Kathryn M.; Slone, Stacey; Gustafson, Alison title: Rural SNAP Participants and Food Insecurity: How Can Communities Leverage Resources to Meet the Growing Food Insecurity Status of Rural and Low-Income Residents? date: 2020-08-19 journal: Int J Environ Res Public Health DOI: 10.3390/ijerph17176037 sha: doc_id: 351785 cord_uid: d35kqobp The burden of obesity disproportionately influences poor health outcomes in rural communities in the United States. Various social and environmental factors contribute to inadequate food access and availability in rural areas, influencing dietary intakes and food insecurity rates. This study aims to identify patterns related to food insecurity and fruit and vegetable consumption within a SNAP-eligible and low-income, highly obese rural Appalachian community. A prospective cohort was implemented to identify gaps in resources addressing obesity and food insecurity challenges. SAS 9.4 software was used to examine differences in dietary intakes and shopping practices among SNAP participants. Among participants (n = 152), most reported an annual household income less than USD 20,000 (n = 90, 60.4%), 29.1% reported food insecurity, and 39.5% reported receiving SNAP benefits within the last month. The overall mean FV intake was 3.46 daily servings (95% CI: 3.06–3.91) among all participants. SNAP participation was associated with food insecurity (p = 0.007) and those participating in SNAP were two times more likely to report being food insecure (OR = 2.707, 95% CI: 1.317, 5.563), relative to non-participants. These findings further depict the need for intervention, as the burden of food insecurity persists. Tailoring health-promoting initiatives to consider rurality and SNAP participation is vital for sustainable success among these populations. The burden of obesity and related chronic diseases disproportionately affects rural communities in the United States (U.S.) more so than their urban counterparts [1] . Theories of social disorganization suggest that the intersection between community structure, such as poverty, socioeconomic status (SES), and residential instability, can result in a void of health promoting culture, infrastructure, and efficacy [2, 3] . Previous insights have shown disparaging differences between urban and rural areas on mortality, chronic disease, and screening rates [3, 4] . Residents' limited knowledge of health promoting behaviors may lead to poor health literacy and unhealthy lifestyle behaviors, including poor dietary intakes and sedentary physical activity levels [5] [6] [7] . Thus, the degree of rurality among geographic areas throughout the U.S. influences the numerous barriers rural communities face and, consequently, their morbidity and mortality rates. Among rural populations, myriad factors affect obesity rates, though fruit and vegetable (FV) intakes are of great influence and few U.S. adults are meeting recommended amounts [8] . This is particularly true in rural communities, where adults exhibit higher obesity prevalence and are less likely to meet daily FV recommendations due to various social and environmental factors [1, 9] relative to their urban counterparts. In addition to individual level factors associated with poor dietary intake, rural residents also face greater rates of food insecurity [10] . A depleted or limited food landscape can predispose residents' dietary consumption and shopping patterns thereby further influencing their health status, as diet is a contributing factor in several chronic illnesses [11] . While agriculture and food production are prominent in many rural landscapes across the U.S., it is not the case for all rural communities. Rurality does not equate to farmland or local food production, which many would think support food security within these communities. Further, the 2017 Census of Agriculture revealed a decline in number of farms and farmers and in acres of farmland and farmland production [12] . At the local level, there are numerous factors that dictate food production, including geography, terrain, and inadequate resources such as economic hardship or lack of farmers. Those who do operate small farms rely on additional off-farm sources for household income [13] . These factors can also influence the household food environment in rural areas. Among low income rural populations, the household food environment, including food security and income concerns, are key factors controlling food choice [14] . Rural communities continue to face higher rates of food insecurity, compared with their urban counterparts [10] , and food insecurity has been associated with obesity and greater cardiometabolic risk [15] . The Supplemental Nutrition Assistance Program (SNAP) is the largest federally funded nutrition program in the U.S., serving as a household-supporting infrastructure for individuals facing food insecurity. SNAP assists eligible, low-income individuals and families in need throughout the U.S. [16] . While eligibility varies by state, those whose income and resources fall below certain thresholds are able to supplement their food budgets using SNAP benefits [16] . Thus, SNAP is often considered a vital resource for those living in rural communities, as the perpetual SES divide continues between rural and urban settings [17] . At the national level, approximately 16% of those living in rural communities live below the federal poverty line, compared with 12% in urban areas [18] . Due to these income gaps, SNAP participation is higher in rural areas, with 16% of households participating, compared with 13% in urban areas [19] . Additionally, most recent federal data from 2017 indicate that of those eligible for SNAP, participation is higher in rural areas (90%) compared to urban areas (82%), and this participation gap continues to climb [20] . Rural areas account for 63% of counties in the U.S., and 87% of counties with the highest rates of food insecurity [21] . Furthermore, a report from 2018 indicates that 13.3% of those living in rural areas faced food insecurity, compared with 11.5% in urban areas [10] . Resources, such as SNAP benefits, and other programs for those of low SES, are imperative for those in rural communities, as many in these areas are at risk of being food insecure. Thus, initiatives like SNAP can aid in alleviating food insecurity among vulnerable households and improve dietary intakes, when adequate access to nutritious choices are available [22] . Community-based efforts have emphasized the importance of looking at social and physical environments when striving to improve food access [23] [24] [25] . Therefore, community-based efforts focused on addressing the local food system are necessary to alleviate the barriers related to the procurement of nutritious foods in rural areas. Prioritizing engagement with key stakeholders and community members is vital to consider how to best approach food access initiatives in rural communities. Conceptually, community-based efforts can be successful in rural communities, as the multifaceted community setting plays a vital role in influencing the food environment and, ultimately, diet choice in these communities. Improving health outcomes pose unique challenges, as resources are sparse and healthcare infrastructure is limited; however, modifying or improving the existing food environment encourages nutritious food choices and shopping behaviors. Nonetheless, environmental triggers and product availability affect the dietary choices individuals make, influencing overall health and obesity status [26] . Given the unique limitations rural communities face, exploring frequented destinations to assess availability can be beneficial to mitigating the barriers that exist [27] . Knowing one's food environment, SNAP participation, and food insecurity status can influence diet quality, an understanding of the interrelationship among these factors can provide guidance for intervention. This study aims to identify patterns related to FV consumption and food access within a SNAP-eligible and low-income, highly obese rural Appalachian county in Kentucky. These findings will serve as a baseline to provide context for addressing food insecurity in a remote rural region of the U.S. Baseline findings will guide points of intercept, design future programming to explore the impact rurality has on obesity status, and address the barriers related to accessing nutritious foods within this community and those similar. The present study is part of a multi-year High Obesity Program (HOP) project through the Centers for Disease Control and Prevention (CDC) to reduce rural obesity prevalence and decrease the risk of chronic disease and preventable mortality. This paper describes one component of the HOP project aimed at providing increased geographic or financial access to nutritious foods. Efforts to improve food access will address food insecurity. This work was completed by leveraging existing Cooperative Extension (CES) infrastructure, with an emphasis placed on community partnership and empowerment, thus enforcing action via established community infrastructure. The CDC funding announcement identified eligible counties across the U.S. based on their obesity prevalence. The setting for this funded project was one eligible Appalachian county in Kentucky with an adult obesity prevalence greater than 40% per the CDC. The Appalachian region of the U.S. has continued to experience significant decline in life expectancy [28] , lack of economic development, and stark out-migration, leaving once fervent and thriving communities destitute, impoverished, and struggling to prosper [29] . This community is reflective of the region, experiencing a persistently high rate of poverty and unemployment, low educational attainment, and food insecurity. The CDC's Social Vulnerability Index, comprised of social and economic indicators, designates the county as "highly vulnerable." [30] The county population is approximately 11,200, and declining, with a median household income of USD $35,000 and an estimated 39% of the population living in poverty [31] . The estimated food insecurity rate is 21%, and approximately 31% of households participate in SNAP [32, 33] . In order to assure broad community input into all program activities, a health coalition was formed, comprised of key stakeholders including local officials (mayor, magistrates), school representatives (food service director, family resource coordinators), library director, concerned citizens, health department representatives, faith-based organization representatives, and community advocates. The health coalition has been pivotal in establishing partnerships to improve health outcomes within the community. It continues to provide input and direction for all aspects of the current project to identify and implement nutrition-related strategies to address the issue of obesity in the county. The current study aims to identify gaps in community resources to establish new partnerships that address obesity and food insecurity challenges. Therefore, a formative food system assessment was conducted at baseline to identify potential areas for intervention to enhance healthier food procurement options. Figure 1 outlines the community's primary food access points identified through the food systems assessment. Findings from the assessment were shared with the coalition to identify potential programmatic efforts to reduce food insecurity within the community. In alignment with the aim of this study, and to complement community efforts, a prospective cohort was enrolled at baseline for a longitudinal study. The prospective cohort study included a face-to-face survey that occurred in year 1 and will again at years 2 and 3. The University of Kentucky Institutional Review Board (IRB) approved the research, promotional materials, consent forms, and survey instrument. In summer 2019, messages on the county's CES Facebook page recruited community residents interested in participating in the cohort study. The CES office, a local food pantry, several faith-based organizations, and grocery stores in the county distributed recruitment materials. Furthermore, recruitment occurred through current community programs offered at the CES office. Recruitment messaging continued until enough individuals enrolled to meet the required sample size, which allowed for attrition. Participants were excluded if they were under 21 years of age, lived outside the county, were non-English speaking, reported plans to move within the next three years, had lived in the county for less than one year, or if they had been diagnosed with cancer. Invited study participants completed the survey via a face-to-face meeting. Prior to survey administration, interviewers verbally reviewed key points of the consent form with participants, who then reviewed the full consent form independently, and provided an opportunity to ask questions or to decline participation. Once deemed eligible, and agreeable to participation, the participant signed the informed consent form and enrolled in the study. A statistical power analysis was performed for sample size estimation and a proposed sample size of 150 adults were recruited to allow for expected attrition. The prospective cohort study surveys were administered at three locations in the county on various days in fall 2019: the CES office, a local food pantry, and the senior citizens center. The initial date of administration had greater turnout than anticipated by study personnel; several surveys were self-administered as a result (n = 24). Moving forward, study personnel modified recruitment processes to schedule appointments for each eligible participant to partake in a verbally administered survey. Those interested contacted study personnel or the county's CES office to schedule a day and time to participate. This resulted in fewer ineligible participants and complete survey responses. The In alignment with the aim of this study, and to complement community efforts, a prospective cohort was enrolled at baseline for a longitudinal study. The prospective cohort study included a face-to-face survey that occurred in year 1 and will again at years 2 and 3. The University of Kentucky Institutional Review Board (IRB) approved the research, promotional materials, consent forms, and survey instrument. In summer 2019, messages on the county's CES Facebook page recruited community residents interested in participating in the cohort study. The CES office, a local food pantry, several faith-based organizations, and grocery stores in the county distributed recruitment materials. Furthermore, recruitment occurred through current community programs offered at the CES office. Recruitment messaging continued until enough individuals enrolled to meet the required sample size, which allowed for attrition. Participants were excluded if they were under 21 years of age, lived outside the county, were non-English speaking, reported plans to move within the next three years, had lived in the county for less than one year, or if they had been diagnosed with cancer. Invited study participants completed the survey via a face-to-face meeting. Prior to survey administration, interviewers verbally reviewed key points of the consent form with participants, who then reviewed the full consent form independently, and provided an opportunity to ask questions or to decline participation. Once deemed eligible, and agreeable to participation, the participant signed the informed consent form and enrolled in the study. A statistical power analysis was performed for sample size estimation and a proposed sample size of 150 adults were recruited to allow for expected attrition. The prospective cohort study surveys were administered at three locations in the county on various days in fall 2019: the CES office, a local food pantry, and the senior citizens center. The initial date of administration had greater turnout than anticipated by study personnel; several surveys were self-administered as a result (n = 24). Moving forward, study personnel modified recruitment processes to schedule appointments for each eligible participant to partake in a verbally administered survey. Those interested contacted study personnel or the county's CES office to schedule a day and time to participate. This resulted in fewer ineligible participants and complete survey responses. The administered survey took approximately 45-60 min to complete. Participants received a USD $25 incentive to be used a local grocery store as compensation for completing the survey. The survey instrument utilized for this cohort comprised a variety of items to measure FV intake, household environmental measures, food purchasing practices, and demographic characteristics. Demographic items in this analysis included age (in years), gender, preferred language, residential status, highest attained education level, race, and annual household income. SNAP participation was assessed by asking: "In the past month, did you or any member of your household receive SNAP benefits or food stamps?" Response options included 'yes' or 'no'. Questions from the National Cancer Institute (NCI) Fruit and Vegetable Intake Screener [34, 35] assessed FV intake. The NCI screener asks respondents about usual intake of various FV, ranging from never to ≥5 times per day, and portion sizes for every item (e.g., "Over the last month, how many times per month, week, or day did you eat fruit?" and "Each time you ate fruit, how much did you usually eat?"). Items include 100% fruit juice, fruit, lettuce salad, French fries or fried potatoes, other white potatoes, cooked dried beans, other vegetables, tomato sauce, vegetable soup, and The survey instrument utilized for this cohort comprised a variety of items to measure FV intake, household environmental measures, food purchasing practices, and demographic characteristics. Demographic items in this analysis included age (in years), gender, preferred language, residential status, highest attained education level, race, and annual household income. SNAP participation was assessed by asking: "In the past month, did you or any member of your household receive SNAP benefits or food stamps?" Response options included 'yes' or 'no'. Questions from the National Cancer Institute (NCI) Fruit and Vegetable Intake Screener [34, 35] assessed FV intake. The NCI screener asks respondents about usual intake of various FV, ranging from never to ≥5 times per day, and portion sizes for every item (e.g., "Over the last month, how many times per month, week, or day did you eat fruit?" and "Each time you ate fruit, how much did you usually eat?"). Items include 100% fruit juice, fruit, lettuce salad, French fries or fried potatoes, other white potatoes, cooked dried beans, other vegetables, tomato sauce, vegetable soup, and mixtures that included vegetables. Summed items created an overall measure of FV intakes among the sample. This measure served as the primary dependent variable for analysis because increased FV intake is a primary goal of the CDC HOP project. The secondary dependent variable, food insecurity, was assessed by asking "Which of the following statements best describes the amount of food eaten in your household in the last 30 days?"-enough food to eat, sometimes not enough to eat, or often not enough to eat [36] . "Sometimes not enough to eat" and "Often not enough to eat" were collapsed into "Not enough food to eat" to create a dichotomous assessment of food insecurity. Potential covariates of interest included gender, income, education, and years of residency. To minimize skewedness, income, education level, and residential status categories were collapsed: income was dichotomized as