The Development, Implementation, and Evaluation of Innovative Strategies to Reduce Food Insecurity among Children in the United States

Abstract

Food insecurity is a health and nutrition concern for many low-income households in the United States. Of particular concern is when food insecurity occurs in households with children, as it is associated with lower fruit consumption, iron-deficient anemia, lower learning and social development, and mental health disorders.1Hanson K.L. Connor L.M. Food insecurity and dietary quality in US adults and children: A systematic review.Am J Clin Nutr. 2014; 100: 684-692Crossref PubMed Scopus (246) Google Scholar, 2Eicher-Miller H.A. Mason A.C. Weaver C.M. McCabe G.P. Boushey C.J. Food insecurity is associated with iron deficiency anemia in US adolescents.Am J Clin Nutr. 2009; 90: 1358-1371Crossref PubMed Scopus (119) Google Scholar, 3Jyoti D.F. Frongillo E.A. Jones S.J. Food insecurity affects school children’s academic performance, weight gain, and social skills.J Nutr. 2005; 135: 2831-2839Crossref PubMed Scopus (535) Google Scholar, 4Burke M.P. Martini L.H. Çayır E. Hartline-Grafton H.L. Meade R.L. Severity of household food insecurity is positively associated with mental disorders among children and adolescents in the United States.J Nutr. 2016; 146: 2019-2026Crossref PubMed Scopus (55) Google Scholar In 2018, among households with children and incomes <185% of the federal poverty level, the rate of household food insecurity was 31% and the rate of food insecurity among children was 16.5%.5Coleman-Jensen A. Rabbitt M.P. Gregory C. Singh A. Household Food Security in the United States in 2018. US Department of Agriculture, Economic Research Services, Washington, DC2019Google Scholar Given the prevalence of food insecurity among low-income households with children, and its association with negative outcomes, policy-makers sought strategies to address the problem. The passage of the Child Nutrition Reauthorization Act of 2010, also referred to as The Healthy, Hunger-Free Kids Act (HHFKA) of 2010,6Healthy, Hunger-Free Kids Act of 2010. Pub L. No. 111-296, 124 Stat 3183 (2010).Google Scholar added a new Section on Childhood Hunger Research to the Richard B. Russell National School Lunch Act.7Richard B. Russell National School Lunch Act. Pub L. No. 396, 60 Stat 230 (1946).Google Scholar This legislation provided $40 million to the US Department of Agriculture (USDA) to conduct and evaluate demonstration projects that test innovative strategies to end childhood hunger, including testing innovative program delivery models and enhanced benefit levels. The legislation required that at least 1 demonstration take place in a rural area on an Indian reservation where the prevalence of diabetes exceeded 15%. It also required an independent evaluation of each demonstration project, using rigorous experimental designs and methodologies, to produce scientifically valid evidence on project impacts on children’s food security. The articles contained in this Supplement to the Journal of the Academy of Nutrition and Dietetics report on the results of 4 demonstration projects that resulted from HHFKA’s legislative mandate.8Burke M.P. Cabili C. Berman D. Forrestal S.G. Gleason P.M. A randomized controlled trial of three school meals and weekend food backpacks on food security in Virginia.J Acad Nutr Diet. 2021; 121: S34-S45Abstract Full Text Full Text PDF PubMed Scopus (4) Google Scholar, 9Briefel R.R. Chojnacki G.J. Gabor V. et al.A cluster randomized controlled trial of a home-delivered food box on food security in Chickasaw Nation.J Acad Nutr Diet. 2021; 121: S46-S58Abstract Full Text Full Text PDF PubMed Scopus (6) Google Scholar, 10Cabili C. Briefel R.R. Forrestal S.G. Gabor V. Chojnacki G.J. A cluster randomized controlled trial of a home-delivered food box on children’s diet quality in the Chickasaw Nation Packed Promise Project.J Acad Nutr Diet. 2021; 121: S59-S69Abstract Full Text Full Text PDF PubMed Scopus (4) Google Scholar, 11Chojnacki G.J. Gothro A.G. Gleason P.M. Forrestal S.G. 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.J Acad Nutr Diet. 2021; 121: S9-S21Abstract Full Text Full Text PDF PubMed Scopus (5) Google Scholar, 12Gleason P.M. Kleinman R.E. Chojnacki G.J. Briefel R. Forrestal S.G. Measuring the effects of a demonstration to reduce childhood food insecurity: A randomized controlled trial of the Nevada Healthy, Hunger Free Kids project.J Acad Nutr Diet. 2021; 121: S22-S33Abstract Full Text Full Text PDF PubMed Scopus (4) Google Scholar Also included in the Supplement are 3 editorials that discuss the findings and the implications they have for food insecurity and childhood hunger research in the United States.13Fram M.S. Frongillo E.A. Moving beyond giving free food: specific targeting and tailoring in response to child food insecurity.J Acad Nutr Diet. 2021; 121: S74-S77Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar, 14Gundersen C. A Consideration of the Evaluation of Demonstration Projects to End Childhood Hunger (EDECH).J Acad Nutr Diet. 2021; 121: S78-S80Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar, 15Eicher-Miller H.A. The need for investment in rigorous interventions to improve child food security.J Acad Nutr Diet. 2021; 121: S70-S73Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar States, the District of Columbia, US territories, and Indian Tribal Organizations (ITOs) that administered federal nutrition assistance programs were eligible to apply to conduct a demonstration project. Proposed projects could include enhanced Supplemental Nutrition Assistance Program (SNAP) benefits for eligible households with children; enhanced benefits or innovative program delivery models in school meals, afterschool snacks programs, and the Child and Adult Care Food Program; and other targeted federal, state or local assistance, including refundable tax credits, emergency housing, employment and training, or family preservation services, for households with children who are experiencing food insecurity. Demonstration sites could not be statewide but were to be located in areas where at least 15% of the households with children had incomes that were below the federal poverty line. From among the proposals received, USDA’s Food and Nutrition Service (FNS)16Food and Nutrition Service, US Department of Agriculture.https://www.fns.usda.govGoogle Scholar competitively selected 3 states (Kentucky, Nevada, and Virginia) and 2 ITOs (Chickasaw Nation and Navajo Nation) in February 2015 to receive cooperative agreements to operate demonstration projects. Both ITOs proposed projects in rural areas where the prevalence of diabetes was at least 15%. Each site was responsible for planning and implementing their demonstration, with oversight from FNS. The competitively selected evaluator, Mathematica, provided technical assistance for the data requirements of the evaluation. An overview of the demonstration projects is presented in the Table. In sum, Chickasaw Nation delivered food boxes and vouchers to purchase fresh or frozen fruits and vegetables to participants’ homes; Kentucky provided additional SNAP benefits based on transportation costs to grocery stores and an earned income deduction; Navajo Nation collaborated with schools and communities to increase availability and enrollment in nutrition assistance programs; Nevada provided extra SNAP benefits to households with young children; and Virginia offered 3 free school meals and take-home food backpacks. It was not possible to conduct a rigorous randomized controlled trial in Navajo Nation because the community-based project did not allow for an unexposed control group. Therefore, the Navajo Nation demonstration was not evaluated, but the demonstration was implemented.Table 1Overview of demonstration projects to end childhood hungerSiteLocationBenefits and servicesTarget populationTime frame (duration)Chickasaw Nation40 school districts (20 treatment, 20 control) in 12 rural counties in OklahomaMonthly home-delivered food boxes containing shelf-stable, nutritious foods and a $15 voucher for fresh or frozen fruits and vegetables per eligible childChildren eligible for free school meals or attending a school where all children receive free school mealsFebruary 2016-February 2018 (25 mo)Kentucky17 rural counties in eastern KentuckyA monthly increase in SNAPaSNAP = Supplemental Nutrition Assistance Program. benefits,bKentucky and Nevada funded and tracked the demonstration grant benefits separately from regular SNAP benefits. Because project participants used the grant benefits as they would their regular SNAP benefits, however, they are referred to as extra SNAP benefits for simplicity. calculated as a fixed income deduction based on the county’s average distance to the grocery store plus a 10% earned income deduction, totaling approximately $45 to $55 per month for households eligible for both deductions, or $9 to $20 per month for households without earned incomeSNAP households with children younger than 18 y and positive net incomeJanuary 2017-March 2018 (15 mo)Navajo Nation3 rural regions in New Mexico and ArizonaCollaboration with schools and communities to increase the availability of and enrollment in nutrition assistance programs based on asset and gap assessmentsChildren younger than 18 ySeptember 2016-June 2018 (22 mo)Nevada12 ZIP codes in Las Vegas$40/mo extra SNAP benefitscForty schools were randomized, but school consolidations and dropouts resulted in 38 schools in the evaluation. per eligible child (treatment group 1), or $40 extra SNAP benefits plus case management and nutrition education (treatment group 2)Households participating in SNAP with incomes <75% of the federal poverty level and at least 1 child younger than 5 yJune 2016-May 2017 (12 mo)Virginia38 schools in rural southwest school districts (10 treatment and 10 control) and Richmond City school district (9 treatment, 9 control)cForty schools were randomized, but school consolidations and dropouts resulted in 38 schools in the evaluation.(1) 3 meals during the school day, (2) food backpacks for weekends and school breaks, (3) monthly summer EBTdEBT = electronic benefit transfer. benefits for each eligible child of $60 in summer 2016 and $30 in summer 2017, and (4) nutrition education for parents and guardiansAll children in treatment schools were offered 3 school meals at school and food backpacks for weekends and school breaks; those eligible for free or reduced-price school meals were also offered summer EBT benefitsJune 2016-June 2018 (24 moeThe evaluation of Virginia’s demonstration focused on the impact of benefits provided during the 2016-2017 school year only.)a SNAP = Supplemental Nutrition Assistance Program.b Kentucky and Nevada funded and tracked the demonstration grant benefits separately from regular SNAP benefits. Because project participants used the grant benefits as they would their regular SNAP benefits, however, they are referred to as extra SNAP benefits for simplicity.c Forty schools were randomized, but school consolidations and dropouts resulted in 38 schools in the evaluation.d EBT = electronic benefit transfer.e The evaluation of Virginia’s demonstration focused on the impact of benefits provided during the 2016-2017 school year only. Open table in a new tab Each evaluation used a randomized controlled trial design to estimate each project’s impact on children’s food security and other outcomes. Either households (Kentucky and Nevada), schools (Virginia), or school districts (Chickasaw Nation) were randomly assigned to a treatment or control group. Households in the treatment group received the demonstration benefits and the control group did not receive the benefits. The evaluation’s primary outcome was food insecurity among children in the previous 30 days. We measured food insecurity among children rather than hunger. Food insecurity is an economic and social condition of limited access to food, and hunger is an individual-level physiological condition that can result from food insecurity. There are currently no valid measures of resource-constrained hunger in the United States. The evaluations were designed to ensure that there was sufficient statistical power to detect a 5-percentage point treatment vs control difference in food insecurity among children. Secondary outcomes included food security among adults and the household as a whole, food spending, and participation in nutrition assistance programs. The evaluation of Chickasaw Nation’s project also focused on children’s food consumption and diet quality because this project was designed to improve household access to healthy food. Data for the impact evaluations were collected through household surveys fielded at baseline (before the demonstration project was implemented) and at follow-up (approximately 12 months after the baseline survey) among both treatment and control households. A second follow-up survey was fielded in Chickasaw Nation approximately 18 months after the baseline survey. Administrative data on measures, such as SNAP participation, SNAP benefit receipt and use, and receipt of other forms of demonstration program assistance, were obtained from states as applicable. Each evaluation used statistical survey weights to help ensure that the results of the demonstration would be representative of the full universe of eligible households in each site and to reduce bias in survey nonresponse. The articles and editorials contained in this Supplement primarily cover the results of the impact evaluations, but the results on all aspects of the demonstrations and their evaluation can be found on the USDA FNS website.17Food and Nutrition Service, US Department of AgricultureEvaluation of Demonstration Projects to End Childhood Hunger.https://www.fns.usda.gov/ops/evaluation-demonstration-projects-echGoogle Scholar Each evaluation measured the food security status of the household as a whole, adults in the household, and children in the household using the 30-day USDA Household Food Security Survey Module (HFSSM),5Coleman-Jensen A. Rabbitt M.P. Gregory C. Singh A. Household Food Security in the United States in 2018. US Department of Agriculture, Economic Research Services, Washington, DC2019Google Scholar which is the reference standard measure in the United States. The module consists of 18 Likert-type items: 10 at the adult level and 8 at the child level. The adult and child items are ordered by severity such that more severe experiences of disruptions in dietary quality and quantity are placed later in the module. It is important to note that in households with multiple adults or children, the adult and child-level food security classifications are for all adults or children. That is, the HFSSM does not measure the specific food security status of each individual in the household; the exception to this rule is for households with either a single adult or a single child. Finally, the evaluations used the HFSSM with a 30-day recall period rather than a 12-month recall period. This was done for 2 key reasons. First, we needed to ensure that the reference period covered only the demonstration period and not the time before the demonstration was delivered. Given the data collection schedule and the timing of the implementation of the demonstrations, only the 30-day measure allowed this to occur. Second, FNS wanted to ensure that the evaluations of the demonstration projects to end childhood hunger were comparable to other high-profile demonstrations, such as the Summer Food for Children Demonstration, which also used the 30-day reference period.18Collins A.M. Klerman J.A. Briefel R. et al.A summer nutrition benefit pilot program and low-income children’s food security.Pediatrics. 2018; 141e20171657Crossref PubMed Scopus (14) Google Scholar HFSSM items are first scored, then summed together. Based on the number of affirmations, children, adults, and households are classified into 1 of 3 categories: Food secure, low food secure, or very low food secure. Children are food secure if 0 to 1 child-level items are affirmed. They are low food secure if 2 to 4 items are affirmed, and very low food secure if 5 to 8 items are affirmed. Adults are food secure if 0 to 2 adult-level items are affirmed. They are low food secure if 3 to 7 items are affirmed and very low food secure if 6 to 10 items are affirmed. Households are food secure if 0 to 2 adult or child items are affirmed. They are low food secure if 3 to 7 items are affirmed and very low food secure if 8 to 18 items are affirmed. Households, adults, or children that are classified as low or very low food secure are collectively known as food-insecure households, adults, or children. More information about the HFSSM and its scoring, use, and reliability can be found on the USDA Economic Research Service website.19Food Security in the US. Accessed October 24, 2020. Economic Research Service, US Department of Agriculture. https://www.ers.usda.gov/topics/food-nutrition-assistance/food-security-in-the-usGoogle Scholar FNS, in partnership with participating demonstration sites and Mathematica, successfully completed the requirements set forth in HHFKA. Although none of the demonstrations significantly reduced the prevalence of food insecurity among children, each one provided important information on interventions that aim to reduce food insecurity among children. For example, the Virginia demonstration reduced very low food insecurity among children, but had negative impacts on other measures of food security. The Chickasaw Nation demonstration reduced food insecurity among adults and modestly increased fruit and vegetable and whole grain consumption in children. In addition, the demonstrations built capacity among states, ITOs, and staff related to developing and implementing large-scale nutrition assistance programs, as evident in the high implementation fidelity of the projects. The articles and editorials in this Supplement offer possible explanations as to why the demonstration projects did not reduce food insecurity as intended. Nevertheless, the results of the evaluations provide valuable information for both policy-makers and researchers to use when considering strategies to reduce food insecurity and end childhood hunger.