Obesity in childhood increases the risk of the leading forms of chronic diseases in adulthood (e.g., cardiometabolic disease and cancer) that cause premature mortality in the U.S. Prevention of childhood life food insecurity and nutritional deficiencies implicated in this cascade of disease, disability and mortality has been the primary objective of the USDA free and reduced price meal (FRPM) programs for over six decades. Unfortunately, there has been increasing (albeit mixed) evidence to suggest that such programs may actually contribute to child obesity in low-income populations. The 2012 Healthy Hunger Free Kids Act (HHFKA) was designed to improve the nutritional quality of school foods and beverages and the Community Eligibility Provision (CEP) of the legislation allows schools serving predominantly low-income populations to offer free meals to all students. The proposed project will be the first to exploit these two federal policies offer to design natural experiments to test hypotheses about the impact of FRPM on food insecurity, child body mass index (BMI) and disparities in child BMI trajectories. The project specific aims are to: 1) use national policy changes in FRPM (i.e., HHFKA and CEP) to better understand how changes in meal nutritional requirements and FRPM participation affects children's food security and BMI trajectories; 2) evaluate whether food insecurity mediates the effect of FRPM on BMI trajectories; and 3) examine the impact of HHFKA and CEP policy changes in FRPM on disparities in Aims 1 and 2 by race/ethnicity and maternal nativity. We achieve these objectives using large nationally representative, ethnically diverse longitudinal data that span pre- and post-HHFKA and CEP implementation from two surveys in the National Center for Education Statistics' Early Childhood Longitudinal Study (ECLS) program, i.e. the ECLS-K:2011, (n=18,200) and the ECLS-K (n= 22,782). Both surveys followed children from kindergarten through at least 5th grade collecting rich data on school meal participation, food insecurity, socioeconomic status and students' measured height and weight. We will use a general panel model estimated using structural equation models for that allows us to flexibly examine and test the assumptions we make in structuring our analyses as natural experiments. We also incorporate novel (multi-level and panel) extensions to propensity score matching to improve balance between `treated' and `untreated' children in these natural experiments. The results are expected to improve our understanding of how school meal programs impact children across school contexts, where access is lacking, and will help inform how federal school meal policies may effectively support health in all children.
We propose to use two nationally representative samples of children, leveraging national policy changes in Healthy Hunger Free Kids Act and the Community Eligibility Provision to better understand how changes in meal nutritional requirements and access to school meals affects children's food security and BMI trajectories. Data will be analyzed using difference in difference and structural modeling approaches to test the impact of these natural experiments on child BMI trajectory indirectly through food insecurity. The results will improve our understanding of how school meal programs impact children across school contexts, where access is lacking, and will help inform how federal school meal policies may effectively support health in all children
