Location: Office of The Area Director
2024 Annual Report
Objectives
1. Determine if the association between self-assessed and measured diet quality (based on the Healthy Eating Index) in adolescents is moderated by gender, body mass index, and food related behaviors using the most recent National Health and Nutrition Examination Survey and associated Flexible Consumer Behavior Survey datasets.
2. Determine if health behavior relationships in families are moderated or mediated by psychosocial constructs related to diet and physical activity, parenting style, and community/neighborhood environments using the Family Life, Activity, Sun, Health and Eating Study datasets.
3. Measure the different food environments to which children living in the Lower Mississippi Delta region are exposed.
Approach
1. Associations between self-assessed and measured diet quality will be determined for adolescents 16-19 years of age using the 2 most recent cycles of the National Health and Nutrition Examination Survey (NHANES). NHANES is a program of studies designed to assess the health and nutritional status of adults and children in the United States. Self-assessed diet quality will be based upon the Flexible Consumer Behavior Survey item “In general, how healthy is your overall diet?”. Measured diet quality will be computed using dietary intake data collected in the dietary interview component of NHANES, and the Healthy Eating Index, a diet quality scoring system. Moderation analysis will be used to determine if the association between self-assessed and measured diet quality is differentially influenced by gender, body mass index, and food related behaviors.
2. Relationships between adolescent and parent/caregiver dietary and physical activity behavior will be determined using the Family Life, Activity, Sun, Health and Eating Study (FLASHE) datasets. FLASHE is a cross-sectional, Internet based survey that was administered to dyads of parents and their adolescent child (aged 12-17 years) in 2014. It was designed to examine lifestyle behaviors, including diet and physical activity, that relate to cancer risk. Dietary behavior, related psychosocial constructs, and community/neighborhood food environment characteristics will be based upon data from the Teen Diet and Parent Diet Surveys which contained a 27-item dietary screener. Physical activity, measured as minutes of moderate-to-vigorous physical activity, and related psychosocial constructs will be based upon data from the Teen Physical Activity and Parent Physical Activity Surveys. Measures of parenting style will be obtained from the Teen Demographic and Parent Demographic Surveys using the 6 items that pertain to the adolescent/parent(s) relationship. The Actor-Partner Interdependence Model will be used to measure the influence that parents have on their adolescent's dietary and physical activity behaviors.
3. Food outlets will be measured using tools designed to assess community nutrition environments. Childcare facilities will be measured using tools designed to assess child nutrition environments specific to such facilities. Geographical analyses will be conducted to determine if patterns of food outlets are clustered, dispersed, or random in the geographic areas of interest. Census tract data will be mapped to explore spatial relationships among food outlets, child care facilities, and population demographic characteristics such as age, race/ethnicity, and poverty status.
Progress Report
This is the final report for Project 6001-51000-004-000D which terminated in April 2024. It has been replaced by new Project 6001-10700-001-000D. Substantial results were realized over the 5 years of this project.
Objective 1: Matching algorithms were created and applied to 2 diet quality measures – self-assessed (perceived) and measured (based on the 2015 Health Eating Index) – to determine if United States (US) adolescents and adults can accurately assess their diet quality. Additionally, factors affecting accuracy of assessment were determined in nationally representative samples of adolescents and adults. Results indicated that only 12% of adolescents and 15% of adults could accurately assess their diet quality with most overrating the quality of their diet. Because food security was one of the few socioeconomic characteristics associated with accurate assessment of diet quality, this work was extended to explore associations among acculturation, diet quality, and accurate assessment of diet quality in Hispanic and Asian Americans. Two proxy acculturation scales were created using nativity/length of US residence, immigration age, and language spoken in household or for dietary recall. Results indicated that accurate assessment of diet quality percentages were low regardless of acculturation level. However, low acculturation appeared protective against negative effects on diet quality typically observed with food insecurity. These findings provide evidence that consumption of traditional Hispanic and Asian diets may have beneficial effects on immigrant’s diet quality, although work is needed to educate all US populations about what constitutes a healthful diet.
Objective 2: Factors that mediated and moderated health behavior relationships in families were identified. First, actor-partner models were used to determine if interdependence relationships between food intake and food liking existed in parent-adolescent dyads. Results indicated that effects of parents’ food preferences on their adolescents’ dietary intake were stronger than adolescent effects on parents’ dietary intake. Second, latent class analysis was used to identify patterns of parent- and child-reported parenting practices for 4 domains – fruit and vegetable consumption, junk food and sugary drinks consumption, physical activity, and sedentary behaviors. Results indicated that distinct patterns of parenting practices existed representing ranges of practice use from high to low. Patterns representing more positive or constructive practices were associated with more beneficial health outcomes. Third, moderation analysis was used to determine if presence of neighborhood food outlets differentially affected relationships between dietary intake and food outlet shopping or meal sources. Results indicated that presence of neighborhood grocery stores and fast-food restaurants affected relationships between meal sources and dietary intake in adolescents but not in parents. Collectively, these findings provide evidence that interventions designed to positively impact health behaviors in families may need to consider whether and which parenting practices are being used, adolescents’ acceptance of these practices, and neighborhood food environment characteristics that affect health behaviors of family members.
Objective 3: Food environments of the Lower Mississippi Delta (LMD) region were measured through the design and implementation of 4 studies. The Delta Food Outlets Study involved mapping and measuring the nutrition environment of 281 food outlets (grocery stores, convenience stores, and restaurants) in 5 rural LMD towns. Results indicated that nutrition scores were universally low for all food outlets and the rural towns contained both food deserts and food swamps, often in overlapping patterns. The Delta Produce Sources Study, a 12-month observational study, involved collecting measures of availability, source, and price of fresh fruits and vegetables sold at 3 farmers markets and in 12 grocery stores located in 5 rural LMD towns. Data were collected on 13 fruits and 32 vegetables. Results indicated that farmers markets had less variability and higher prices, but more locally sourced produce than grocery stores. The Farm to School (F2S) Procurement and Simulation Study was conducted to determine produce amounts purchased by 2 LMD school districts and to simulate produce amounts needed to implement a locally grown F2S menu. Procurement data from 2 academic years were coded, classified into 8 major food categories, and compared. Results indicated that the school district participating in the USDA Fresh Fruit and Vegetable Program procured a wider variety of fruits and vegetables than did the nonparticipating school district. However, few items were procured locally for either school district. The Mississippi F2S Study involved using online surveys to gauge small farmers’ and school food service directors’ (SFSD) interest in and experience with F2S. The small farmer survey also collected data about type, quantity, and seasonality of food grown, raised, or produced. The SFSD survey also collected data about purchasing practices. In total, 38 small farmers and 122 SFSD completed the surveys. Summary statistics were computed, and geospatial analysis was conducted to determine distances between the small farms and schools. Results indicated that only 11% of small farmers sold to schools although 68% expressed interest in F2S activities. Additionally, 46% of SFSD did not purchase any locally sourced foods, although 64% expressed interest in F2S activities. Most farm produce was harvested during the school year and 98% of farms were within 50-mile radius of at least 1 school. Collectively, these findings provide evidence that the poor nutrition environments of the LMD continue to make it difficult for residents to purchase and consume healthful foods.
Additional studies: An existing food and nutrition security dataset was analyzed to (1) quantify and compare causal pathways from food and nutrition security to dietary choice while testing for mediation by food availability and barriers to healthful meals; (2) compare socioeconomic, dietary choice, and nutrition environment variables associated with food and nutrition security in households with and without children; and (3) identify significant sociodemographic/socioeconomic, chronic disease, dietary habits, and food environment explanatory variables for perceptions of self-rated health. Results from causal mediation analysis indicated that food and nutrition security affect food choices with barriers to healthful meals acting as an intermediary step. Further, when environmental and household utilization barriers to healthful food purchasing and preparation are high, the ability to decide what to eat does not differ between households with nutrition security and those with nutrition insecurity. Results from the logistic regression analyses indicated that socioeconomic and nutrition environment variables differentially affected food and nutrition security in households with and without children. However, dietary choice was an influential factor for food and nutrition security in both household types. Lastly, perceptions of one’s health are positively associated with healthful dietary habits and negatively associated with nutrition insecurity and presence of chronic disease.
The overall impact of these projects included new or complimentary information that researchers, policy makers, and other stakeholders can use when designing studies or implementing policies and programs to improve nutrition and food environments of marginalized and health disparate communities and populations, like those with food and nutrition insecurity and rural Mississippi Delta residents.
Accomplishments
1. Deciding what to eat does not differ between households with and without nutrition security when barriers are high. Nutrition security, or the ability to acquire food that promotes health, is a relatively new concept distinct from food security. Because nutrition security is a relatively new concept, it requires careful evaluation to determine its usefulness for nutrition research. ARS researchers in Stoneville, Mississippi, and University of Central Arkansas collaborators determined that both food and nutrition security affect food choice. However, when environmental and household utilization barriers to healthful food purchasing and preparation are high, the ability to decide what to eat does not differ between households with nutrition security and those without nutrition security. Strategies and policies that remove household utilization barriers may help lessen the burdens of both food and nutrition insecurity.
2. Nutrition insecurity doubles the risk for low self-rated health. Self-rated health is consistently associated with death and disease. While food security has been associated with self-rated health, less is known about relationships between nutrition security and self-rated health. ARS researchers in Stoneville, Mississippi, and University of Central Arkansas collaborators determined that higher household income, more fruit and vegetable intake, and consumption of scratch-cooked meals protected against low self-rated health. In contrast, nutrition insecurity (but not food security) and having at least 1 chronic disease increased risk for low self-rated health. These findings suggest that efforts to improve nutrition security also may improve self-rated health and ultimately overall health and well-being, particularly in populations disproportionately burdened by chronic disease.
Review Publications
Thomson, J.L., Landry, A.S., Walls, T.I. 2024. Direct and indirect effects of food and nutrition security on dietary choice and healthfulness of food choice: causal mediation analysis. Current Developments in Nutrition. 8(2):102081. https://doi.org/10.1016/j.cdnut.2024.102081.
Thomson, J.L., Landry, A., Walls, T.I. 2024. Differences in socioeconomic, dietary choice, and nutrition environment explanatory variables for food and nutrition security among households with and without children. Nutrients. 16(6):883. https://doi.org/10.3390/nu16060883.