Application of Path Analysis and Structural Equation Modeling in Nutrition and Dietetics.

This is one of a series of monographs on research design and analysis. The purpose of this article is to describe a set of statistical procedures or techniques used to develop and test structural models that characterize the relationships and interrelationships between a group of concepts and variables. These procedures include multiple regression, exploratory and confirmatory factor analysis, path analysis, and structural equation modeling. The article describes the purpose of each of these procedures and how they relate to and build on one another. It also covers the different types of variables examined, including the distinction between endogenous, exogenous, and mediating variables, along with the distinction between measured and unmeasured (or latent) variables. Each procedure results in a set of statistical estimates, and the article presents the interpretation of these estimates, including regression coefficients (standardized and unstandardized), path coefficients, factor loadings, and coefficients of determination (or R2 values). The article presents examples of how each procedure has been used in practice, along with additional resources for readers who wish to learn more.

Measuring the Effects of a Demonstration to Reduce Childhood Food Insecurity: A Randomized Controlled Trial of the Nevada Healthy, Hunger Free Kids Project.

BACKGROUND: To reduce childhood hunger, the US Department of Agriculture funded a set of demonstration projects, including the Nevada Healthy, Hunger-Free Kids (HHFK) project. OBJECTIVE: The study objective was to test whether the Nevada HHFK project reduced child food insecurity (FI-C) among low-income households with young children. DESIGN: Households were randomly assigned to treatment and control groups, with outcomes measured using household surveys and administrative data. Survey data were collected at baseline (n=3,088) and follow-up (n=2,074) 8 to 12 months into the project. PARTICIPANTS/SETTING: Eligible households in Las Vegas, NV, had children under age 5 years, received Supplemental Nutrition Assistance Program (SNAP) benefits, and had incomes below 75% of the federal poverty level. INTERVENTION: Between June 2016 and May 2017, treatment households on SNAP received an additional $40 in monthly SNAP benefits per child under age 5 years. MAIN OUTCOME MEASURES: Key outcomes included FI-C (primary), food security among adults and households, and food expenditures (secondary). STATISTICAL ANALYSES PERFORMED: Differences between the treatment and control groups were estimated by a logistic regression model and controlling for baseline characteristics. Analyses were also performed on socioeconomic subgroups. RESULTS: The Nevada HHFK project did not reduce FI-C (treatment=31.2%, control=30.6%; P=0.620), very low food security among children (P=0.915), or food insecurity among adults (P=0.925). The project increased households' monthly food expenditures (including SNAP and out-of-pocket food purchases) by $23 (P<0.001). CONCLUSIONS: A demonstration project to reduce FI-C by increasing SNAP benefits to Las Vegas households with young children and very low income did not reduce FI-C or other food-insecurity measures. This finding runs counter to prior research showing that SNAP and similar forms of food assistance have reduced food insecurity. This project was implemented during a period of substantial economic growth in Las Vegas. Future research should explore the role of the economic context, children's ages, and household income in determining how increases in SNAP benefits affect food insecurity. CLINICALTRIALS. GOV IDENTIFIER: NCT04253743 (http://www.clinicaltrials.gov) FUNDING/SUPPORT: This article is published as part of a supplement supported by the US Department of Agriculture, Food and Nutrition Service.

A Cluster Randomized Controlled Trial of a Home-Delivered Food Box on Food Security in Chickasaw Nation.

BACKGROUND: The 2010 Child Nutrition reauthorization called for the independent evaluation of innovative strategies to reduce the risk of childhood hunger or improve the food security status of households with children. OBJECTIVE: The research question was whether the Packed Promise intervention reduces child food insecurity (FI-C) among low-income households with children. DESIGN: This study was a cluster randomized controlled trial of 40 school districts and 4,750 eligible, consented households within treatment and control schools. PARTICIPANTS/SETTING: Data were collected at baseline (n = 2,859) and 2 follow-ups (n = 2,852; n = 2,790) from households with children eligible for free school meals in participating schools in 12 rural counties within the Chickasaw Nation territory in south central Oklahoma in 2016 to 2018. INTERVENTION: Each month of the 25-month intervention, for each eligible child, enrolled households could choose from 5 types of food boxes that contained shelf-stable, nutritious foods ($38 food value) and a $15 check for purchasing fruits and vegetables. MAIN OUTCOME MEASURES: The primary outcome was FI-C. Other outcomes included household and adult food security, very low food security among children, and food expenditures. STATISTICAL ANALYSES PERFORMED: Differences between the treatment and control groups were estimated by a regression model controlling for baseline characteristics. RESULTS: The Packed Promise project did not significantly reduce FI-C at 12 months (29.3% prevalence in the treatment group compared with 30.1% in the control group; P = 0.123) or at 18 months (28.2% vs 28.7%; P = 0.276), but reduced food insecurity for adults by 3 percentage points at 12 months (P = 0.002) but not at 18 months (P = 0.354). The intervention led to a $27 and a $16 decline in median household monthly out-of-pocket food expenditures at 12 and 18 months, respectively. CONCLUSIONS: An innovative intervention successfully delivered nutritious food boxes to low-income households with children in rural Oklahoma, but did not significantly reduce FI-C. Improving economic conditions in the demonstration area and participation in other nutrition assistance programs among treatment and control groups might explain the lack of impact.ClinicalTrials.gov ID: NCT04316819 (http://www.clinicaltrials.gov). FUNDING/SUPPORT: This article is published as part of a supplement supported by the US Department of Agriculture, Food and Nutrition Service.

A Randomized Controlled Trial Measuring Effects of Extra Supplemental Nutrition Assistance Program (SNAP) Benefits on Child Food Security in Low-Income Families in Rural Kentucky.

BACKGROUND: To reduce childhood hunger, the US Department of Agriculture funded several innovative demonstration projects, including the Kentucky Ticket to Healthy Food project. OBJECTIVE: The study tested the hypothesis that Ticket to Healthy Food would reduce child food insecurity (FI-C) among rural, low-income households. DESIGN: The study used a randomized controlled trial in which households were randomly assigned to treatment and control groups. Outcomes were measured using household surveys and administrative data. Survey data were collected at baseline (n=2,202) and follow-up (n=1,639) 8 to 11 months into the project. PARTICIPANTS/SETTING: Households in 17 counties in southeastern Kentucky that had at least 1 child younger than 18 years and received a Supplemental Nutrition Assistance Program (SNAP) benefit amount less than the maximum at baseline. INTERVENTION: Between January 2017 and March 2018, treatment households on SNAP received additional monthly benefits ranging from $1 to $122 based on distance to grocery store and earned income. MAIN OUTCOME MEASURES: Key outcomes included FI-C (primary), food insecurity among adults and households, and food expenditures (secondary). STATISTICAL ANALYSES PERFORMED: Logistic and linear regression models were used to estimate differences between the treatment and control groups, controlling for baseline characteristics. Socioeconomic subgroups were also analyzed. RESULTS: The Kentucky Ticket to Healthy Food project did not reduce the primary outcome, FI-C (treatment=37.1%, control=35.2%; P=0.812), or secondary outcomes of very low food security among children (treatment=3.7%, control=4.4%; P=0.204) or food insecurity among adults (treatment=53.9%, control=53.0%; P=0.654). The project increased households' monthly food spending by $20 (P=0.030) and led more households to report that monthly benefits lasted at least 3 weeks (treatment=65%, control=56%; P=0.009). CONCLUSION: A demonstration project to reduce FI-C by raising SNAP benefits for Kentucky households with children did not reduce FI-C or other food insecurity measures. Future research should explore the effect of different increases in SNAP benefits and collect repeated measures of FI-C to assess whether intervention effects change over time. FUNDING/SUPPORT: This article is published as part of a supplement supported by the US Department of Agriculture, Food and Nutrition Service.

The Bayesian Approach to Decision Making and Analysis in Nutrition Research and Practice.

This is part of a series of monographs on research design and analysis. The purpose of this article is to describe the purposes of and approach to conducting Bayesian decision making and analysis. Bayesian decision making involves basing decisions on the probability of a successful outcome, where this probability is informed by both prior information and new evidence the decision maker obtains. The statistical analysis that underlies the calculation of these probabilities is Bayesian analysis. In recent years, the Bayesian approach has been applied more commonly in both nutrition research and clinical decision making, and registered dietitian nutritionists would benefit from gaining a deeper understanding of this approach. This article provides a background of Bayesian decision making and analysis, and then presents applications of the approach in two different areas-medical diagnoses and nutrition policy research. It concludes with a description of how Bayesian decision making may be used in everyday life to allow each of us to appropriately weigh established beliefs and prior knowledge with new data and information in order to make well-informed and wise decisions.

What is translational research? Concepts and applications in nutrition and dietetics.

This monograph is tenth in a series of articles focused on research design and analysis, and provides an overview of translational research concepts. Specifically, this article presents models and processes describing translational research, defines key terms, discusses methodological considerations for speeding the translation of nutrition research into practice, illustrates application of translational research concepts for nutrition practitioners and researchers, and provides examples of translational research resources and training opportunities. To promote the efficiency and translation of evidence-based nutrition guidelines into routine clinical-, community-, and policy-based practice, the dissemination and implementation phases of translational research are highlighted and illustrated in this monograph.

Publishing nutrition research: a review of multivariate techniques--part 3: data reduction methods.

This is the ninth in a series of monographs on research design and analysis, and the third in a set of these monographs devoted to multivariate methods. The purpose of this article is to provide an overview of data reduction methods, including principal components analysis, factor analysis, reduced rank regression, and cluster analysis. In the field of nutrition, data reduction methods can be used for three general purposes: for descriptive analysis in which large sets of variables are efficiently summarized, to create variables to be used in subsequent analysis and hypothesis testing, and in questionnaire development. The article describes the situations in which these data reduction methods can be most useful, briefly describes how the underlying statistical analyses are performed, and summarizes how the results of these data reduction methods should be interpreted.

Publishing nutrition research: a review of multivariate techniques--part 2: analysis of variance.

This article is the eighth in a series exploring the importance of research design, statistical analysis, and epidemiology in nutrition and dietetics research, and the second in a series focused on multivariate statistical analytical techniques. The purpose of this review is to examine the statistical technique, analysis of variance (ANOVA), from its simplest to multivariate applications. Many dietetics practitioners are familiar with basic ANOVA, but less informed of the multivariate applications such as multiway ANOVA, repeated-measures ANOVA, analysis of covariance, multiple ANOVA, and multiple analysis of covariance. The article addresses all these applications and includes hypothetical and real examples from the field of dietetics.

Publishing nutrition research: validity, reliability, and diagnostic test assessment in nutrition-related research.

This is the sixth in a series of monographs on research design and analysis. The purpose of this article is to describe and discuss several concepts related to the measurement of nutrition-related characteristics and outcomes, including validity, reliability, and diagnostic tests. The article reviews the methodologic issues related to capturing the various aspects of a given nutrition measure's reliability, including test-retest, inter-item, and interobserver or inter-rater reliability. Similarly, it covers content validity, indicators of absolute vs relative validity, and internal vs external validity. With respect to diagnostic assessment, the article summarizes the concepts of sensitivity and specificity. The hope is that dietetics practitioners will be able to both use high-quality measures of nutrition concepts in their research and recognize these measures in research completed by others.

Association between school food environment and practices and body mass index of US public school children.

BACKGROUND: With the ongoing interest in implementing school policies to address the problem of childhood obesity, there is a need for information about the relationships between school food environments and practices and children's weight status. OBJECTIVE: To examine the association between school food environments and practices and children's body mass index (BMI; calculated as kg/m(2)). DESIGN: The study used data from the third School Nutrition Dietary Assessment Study, a cross-sectional study that included a national sample of public school districts, schools, and children in the 2004-2005 school year. Data on school food environments and practices were collected through on-site observations and interviews with school principals, and children were weighed and measured by trained data collectors. SUBJECTS/SETTING: The study included 287 schools and 2,228 children in grades 1 through 12. STATISTICAL ANALYSES PERFORMED: Ordinary least squares regression was used to estimate the associations between school food environments and practices and BMI z scores and logistic regression was used to estimate associations between school food environments and practices and the likelihood of obesity (defined as BMI-for-age >or=95th percentile). Models included controls for sociodemographic characteristics of schools and children, to control for potential endogeneity of school environments and practices, as well as controls for children's dietary and physical activity behaviors outside of school. RESULTS: Among elementary school children, offering french fries and similar potato products in subsidized school meals more than once per week and offering dessert more than once per week were each associated with a significantly higher likelihood of obesity. Among middle school children, the availability of low-nutrient, energy-dense foods in vending machines in or near the foodservice area was associated with a higher BMI z score, and the availability of such foods for à la carte purchase in the cafeteria was associated with a lower BMI z score. CONCLUSIONS: Findings from this analysis suggest that limiting children's access to low-nutrient, energy-dense foods at school may hold promise as a tactic for reducing children's total calorie intake and controlling children's BMI.

School breakfast program but not school lunch program participation is associated with lower body mass index.

BACKGROUND: Rates of overweight and obesity have increased dramatically during the past 2 decades. Children obtain a large fraction of their food energy while at school. OBJECTIVE: To estimate the relationship between participation in school meal programs and children's body mass index (BMI) and their likelihood of being overweight or obese, testing the hypothesis that school meal participation influences students' weight status, as measured by their BMI and indicators of overweight and obesity. DESIGN: A cross-sectional design in which a regression model was used to estimate the association between participation in the School Breakfast Program and National School Lunch Program and children's BMI and risk of overweight or obesity, controlling for a wide range of student and school characteristics. SUBJECTS/SETTING: Participants included a nationally representative sample from the third School Nutrition Dietary Assessment Study of 2,228 students in grades 1 through 12 for whom height and weight measurements were obtained. These students, along with their parents, each completed a survey. STATISTICAL ANALYSES PERFORMED: Multivariate regression models were used to examine the relationship between usual school meal participation and BMI and indicators of whether students were overweight or obese. These models controlled for students' demographic and socioeconomic characteristics, levels of physical activity, usual eating habits, screen time, and school characteristics. RESULTS: No evidence was found of any relationship between usual school lunch participation and any of four different measures of weight status based on students' BMI. School breakfast participation was associated with significantly lower BMI, particularly among non-Hispanic, white students. CONCLUSIONS: There was no evidence that either the school breakfast or lunch program is contributing to rising rates of childhood obesity. In fact, School Breakfast Program participation may be a protective factor, by encouraging students to consume breakfast more regularly.

School food environments and practices affect dietary behaviors of US public school children.

BACKGROUND: Changes to school food environments and practices that lead to improved dietary behavior are a powerful strategy to reverse the childhood obesity epidemic. OBJECTIVES: To estimate the effects of school food environments and practices, characterized by access to competitive foods and beverages, school lunches, and nutrition promotion, on children's consumption of sugar-sweetened beverages, low-nutrient energy-dense foods, and fruits/vegetables at school. DESIGN: Cross-sectional study using data from the third School Nutrition Dietary Assessment Study, a nationally representative sample of public school districts, schools, and children in school year 2004-2005. Data from school principals and foodservice directors, school menu analysis, and on-site observations were used to characterize school food environments and practices. Dietary intake was assessed using 24-hour recalls. SUBJECTS/SETTING: The sample consists of 287 schools and 2,314 children in grades one through 12. STATISTICAL ANALYSES PERFORMED: Ordinary least squares regression was used to identify the association between school food environments and practices (within elementary, middle, and high schools) and dietary outcomes, controlling for other school and child/family characteristics. RESULTS: Sugar-sweetened beverages obtained at school contributed a daily mean of 29 kcal in middle school children and 46 kcal in high school children across all school children. Attending a school without stores or snack bars was estimated to reduce sugar-sweetened beverage consumption by 22 kcal per school day in middle school children (P<0.01) and by 28 kcal in high school children (P<0.01). The lack of a pouring rights contract in a school reduced sugar-sweetened beverage consumption by 16 kcal (P<0.05), and no à la carte offerings in a school reduced consumption by 52 kcal (P<0.001) in middle school children. The most effective practices for reducing energy from low-energy, energy-dense foods were characteristics of the school meal program; not offering french fries reduced low-nutrient, energy-dense foods consumption by 43 kcal in elementary school children (P<0.01) and sugar-sweetened beverage consumption by 41 kcal in high school children (P<0.001). CONCLUSIONS: To improve children's diet and reduce obesity continued changes to school food environments and practices are essential. Removing sugar-sweetened beverages from school food stores and snack bars, improving à la carte choices, and reducing the frequency of offering french fries merit testing as strategies to reduce energy from low-nutrient, energy-dense foods at school.

Consumption of low-nutrient, energy-dense foods and beverages at school, home, and other locations among school lunch participants and nonparticipants.

BACKGROUND: Access to foods and beverages on school campuses, at home, and other locations affects children's diet quality, energy intake, and risk of obesity. OBJECTIVES: To describe patterns of consumption of "empty calories"--low-nutrient, energy-dense foods, including sugar-sweetened beverages--by eating location among National School Lunch Program (NSLP) participants and nonparticipants. DESIGN: Cross-sectional study using 24-hour dietary recall data from the 2004-2005 third School Nutrition Dietary Assessment Study. SUBJECTS/SETTING: A nationally representative sample of 2,314 children in grades one through 12, including 1,386 NSLP participants. STATISTICAL ANALYSES PERFORMED: Comparisons, using t tests, of the proportion of children consuming low-nutrient, energy-dense foods and beverages, mean daily energy and energy from low-nutrient, energy-dense foods, and energy density by NSLP participation status. RESULTS: On a typical school day, children consumed 527 "empty calories" during a 24-hour period. Eating at home provided the highest mean amount of energy from low-nutrient, energy-dense foods (276 kcal vs 174 kcal at school and 78 kcal at other locations). NSLP participants consumed less energy from sugar-sweetened beverages at school than nonparticipants (11 kcal vs 39 kcal in elementary schools and 45 kcal vs 61 kcal in secondary schools, P<0.01), but more energy from low-nutrient, energy-dense solid foods such as french fries and higher-fat baked goods in secondary schools (157 kcal vs 127 kcal, P<0.01). Participants were not more likely to consume sugar-sweetened beverages or low-nutrient, energy-dense foods at home or other locations. School lunch participants' consumption at school was less energy-dense than nonparticipants' consumption at school (P<0.01). Energy density was highest for consumption at locations away from home and school. CONCLUSIONS: Improving home eating behaviors, where the largest proportion of total daily and energy from low-nutrient, energy-dense foods are consumed (especially from sugar-sweetened beverages, chips, and baked goods) is warranted. At schools, consumption of energy from low-nutrient, energy-dense foods may be reduced by limiting access to competitive foods and beverages, enforcing strong school wellness policies, and minimizing the frequency of offering french fries and similar potato products and higher-fat baked goods in school meals or à la carte.

An introduction to qualitative research for food and nutrition professionals.

The purpose of this article is to define qualitative research, explain its design, explore its congruence with quantitative research, and provide examples of its applications in dietetics. Also, methods to ensure validity, reliability, and relevance are addressed. Readers will gain increased knowledge about qualitative research and greater competency in evaluating this type of research. The hope is that food and nutrition professionals will be inspired to conduct and publish qualitative research, adding to the body of peer-reviewed dietetics-related qualitative publications. This type of research must be methodically planned and implemented with attention to validity, reliability, and relevance. This rigorous approach boosts the probability that the research will add to the scientific literature and qualify for publication.

Using Dietary Reference Intake-based methods to estimate the prevalence of inadequate nutrient intake among school-aged children.

OBJECTIVE: To estimate the prevalence of inadequate usual intakes of nutrients by school-aged children. DESIGN: A descriptive study using data from the US Department of Agriculture 1994 to 1996 Continuing Survey of Food Intake by Individuals. Each subject provided two 24-hour recalls. We adjusted for day-to-day variation in nutrient intake and estimated the percentage of children with intakes below the Estimated Average Requirement (EAR) using the Software for Intake Distribution Estimation Program. SUBJECTS: A national sample of noninstitutionalized children aged 6 to 18 years (N=2,692). STATISTICAL ANALYSES: Chi2 tests showed that background characteristics or percentages with intakes below the EAR were the same across the 6 gender-age and racial/ethnic groups. The SUDAAN statistical package was used to account for the complex sample design. RESULTS: Usual intakes were more favorable for 5 B vitamins and iron than for the other nutrients examined. High percentages of children had intakes below the EAR for vitamin E. Many children aged 9 years and older had intakes below the EAR for folate and magnesium. Females aged 9 years and older had low calcium intakes relative to the Adequate Intake value. Females aged 14 to 18 years were at highest risk of usual intakes that did not meet the EARs. Few males in this age group met the EAR for vitamin E or magnesium. APPLICATIONS: Females aged 14 to 18, in particular, should be targeted for efforts to improve nutrient intakes. Studies should monitor children's usual nutrient intakes after adjusting for day-to-day variation.