Nourishing the Next Generation: Examining How U.S. Policies and Environments Shape Food Security, Youth Diet Quality, and Eating Behaviors

Abstract

Food security, youth diet quality, and a positive relationship with food free from disordered eating behaviors are essential components of optimal health and well-being. We are far from this reality in the United States; over 1 in 6 households with children experience food insecurity and do not have enough food to reliably feed all members of the household, most children’s diet quality is categorized as suboptimal, and over 1 in 5 children and adolescents experience disordered eating behaviors. Population-level policies and programs designed to improve these outcomes are, at times, implemented in such a way that make it difficult for individuals who are in need to access them. Additionally, many policies and programs do not consider additional factors explaining the outcomes, and few, if any, have considered improving these three outcomes in tandem when designing policies and programs or, at the very least, ensuring that unintended consequences are not introduced (i.e. increasing disordered eating behaviors in the process of improving diet quality). In this dissertation, I leverage three distinct datasets and methods to explore gaps and potential solutions to improve food security, youth diet quality, and eating behaviors by 1) evaluating how statewide universal free school meals policies influence food insecurity among families with school-aged children, 2) exploring how parents navigate feeding young children in the American food environment when they are considered responsible for ensuring their children adhere to nutritional recommendations, and 3) understanding patterns of onset of disordered eating behaviors among undergraduate students during college. In Chapter One, we evaluate the association of universal school meals (USM) policies on household- and child-level food insecurity using nationally representative data from the US Census Current Population Survey from 2012-2024. Using a difference-in-differences analytic approach, we find that the highest income households living in USM-maintaining states experienced a 1.3 percentage-point (95% CI: -2.1, -0.5) differential decrease in household food insecurity and a 0.9 percentage-point (95% CI: -1.4, -0.5) differential decrease in child food insecurity compared to households living in states that discontinued the USM policy after the pandemic-related federal emergency waiver. Households with low-middle incomes in USM-maintaining states experienced a 3.2 percentage-point (95% CI: -6.1, -0.3) differential decrease in household food insecurity but no differential change in child food insecurity. Our results show that households that traditionally do not qualify for free meals may benefit from free meals for all children. In Chapter Two, we explore how parents navigate feeding young children within the American food environment, uncovering the joys and challenges of the process using qualitative methods. Guided by the Social Ecological Model, Family Ecological Model, and Stangl’s Health Stigma and Discrimination Framework, we conducted semi-structured interviews with 29 parents of young children 2-10 years old in Massachusetts and used a team-based reflexive thematic analytic approach to develop and identify themes. We found that 1) parents receive a multitude of sometimes-conflicting messages about feeding children, 2) parents develop varying motivations, priorities, and goals that inform decisions when feeding children, and 3) parents confront reality that the process is often different than what they had hoped and, through this, experience an emotional burden. In Chapter Three, we investigate patterns of onset of disordered eating behaviors among undergraduate students during college using a unique longitudinal dataset from four private universities in the southeast US. Using survival analytic techniques, we estimate the probability of onset for binge eating behaviors and disordered weight control behaviors during college and examine sociodemographic variables associated with onset. At baseline, 27% reported current disordered weight control behaviors, 7% of students reported current binge eating behaviors, and 4% reported past history of DEBs prior to college. Among those who did not report past history of DEBs or current disordered weight control behavior at baseline, the onset probability of disordered weight control behavior by spring of their second year was 14.2% (95% CI: 11.4, 16.9) and the cumulative onset probability by graduation 40.3% (95% CI: 36.1, 44.3). Among those who did not report past history of DEBs or current binge eating behavior at baseline, the onset probability of binge eating behavior by spring of their second year was 5.4% (95% CI: 3.8, 7.0) and the cumulative onset probability by graduation was 13.2% (95% CI: 10.7, 15.6). Together, these three chapters explore how population-level programs and policies can further be expanded or developed to support youth and families for improved health and well-being: expand implementation of universal free school meals policies; consider how parents carry an emotional burden when feeding children within a food environment heavily influenced by commercial actors; and implement evidence-based, universal programming for prevention, screening, and treatment of disordered eating behaviors at colleges and universities to mitigate DEB incidence and severity.